Energy Fuels Inc. Technical Report – Donald Rare Earths and Mineral Sands Project, Victoria, Australia

This report was prepared as a National Instrument 43-101 Standards of Disclosure for Mineral Projects (NI 43-101) Technical Report and a Subpart 229.1300 of Regulation S-K Disclosure by Registrants Engaged in Mining Operations (S-K 1300) and Item 601 (b)(96) Technical Report Summary (collectively Technical Report) for Energy Fuels Inc. (Energy Fuels) by Datamine Australia Pty Ltd (Snowden Optiro). The quality of information, conclusions, and estimates contained herein are consistent with the quality of effort involved in Snowden Optiro's services. This Technical Report is intended to satisfy the requirements of a Feasibility Study under both NI 43-101 and S-K 1300. The information, conclusions, and estimates contained herein are based on: i) information available at the time of preparation, ii) data supplied by outside sources, and iii) the assumptions, conditions, and qualifications set forth in this report. The user of this document should ensure that this is the most recent Technical Report for the property as it is not valid if a new Technical Report has been issued.

© 2026

All rights are reserved.

Energy Fuels Inc. Technical Report – Donald Rare Earths and Mineral Sands Project, Victoria, Australia

Table of contents

Energy Fuels Inc. Technical Report – Donald Rare Earths and Mineral Sands Project, Victoria, Australia

Energy Fuels Inc. Technical Report – Donald Rare Earths and Mineral Sands Project, Victoria, Australia

Energy Fuels Inc. Technical Report – Donald Rare Earths and Mineral Sands Project, Victoria, Australia

Energy Fuels Inc. Technical Report – Donald Rare Earths and Mineral Sands Project, Victoria, Australia

Energy Fuels Inc. Technical Report – Donald Rare Earths and Mineral Sands Project, Victoria, Australia

Energy Fuels Inc. Technical Report – Donald Rare Earths and Mineral Sands Project, Victoria, Australia

Figures

Energy Fuels Inc. Technical Report – Donald Rare Earths and Mineral Sands Project, Victoria, Australia

Tables

Energy Fuels Inc. Technical Report – Donald Rare Earths and Mineral Sands Project, Victoria, Australia

Energy Fuels Inc. Technical Report – Donald Rare Earths and Mineral Sands Project, Victoria, Australia

Forward-looking information

This Technical Report contains "forward-looking information" and "forward-looking statements" within the meaning of applicable Canadian and United States securities legislation (collectively, "forward-looking information") which involves a number of risks and uncertainties. Forward-looking information includes, but is not limited to: information with respect to strategy, plans, expectations or future financial or operating performance, such as expectations and guidance regarding project development, production outlook, including estimates of production, grades, recoveries and costs; estimates of Mineral Resources and Mineral Reserves; construction plans; mining and recovery methods; mining and mineral processing and rates; tailings disposal design and capacity; mine life; timing and success of exploration programs and project related risks as well as any other information that expresses plans and expectations or estimates of future performance. Often, but not always, forward-looking information can be identified by the use of words such as "plans", "expects", or "does not expect", "is expected", "budget", "scheduled", "estimates", "forecasts", "intends", "anticipates", or "does not anticipate", or "believes", or variations of such words and phrases or state that certain actions, events or results "may", "could", "would", "might" or "will" be taken, occur or be achieved.

Forward-looking information is based on the opinions, estimates and assumptions of contributors to this Technical Report. Certain key assumptions are discussed in more detail. Forward-looking information involves known and unknown risks, uncertainties and other factors which may cause the actual results, performance or achievements to be materially different from any other future results, performance or achievements expressed or implied by the forward-looking information.

Such factors and assumptions underlying the forward-looking information in this Technical Report includes, but are not limited to: risks associated with community relationships; risks related to estimates of production, cash flows and costs; risks inherent to mining operations; shortages of critical supplies; the cost of non-compliance and compliance; volatility in commodity prices; risks related to compliance with environmental laws and liability for environmental contamination; the lack of availability of infrastructure; risks related to the ability to obtain, maintain or renew regulatory approvals, permits and licenses; imprecision of Mineral Reserve and Mineral Resource estimates; deficient or vulnerable title to concessions, easements and surface rights; inherent safety hazards and risk to the health and safety of employees and contractors; risks related to the workforce and its labour relations; key talent recruitment and retention of key personnel; the adequacy of insurance; uncertainty as to reclamation and decommissioning; the uncertainty regarding risks posed by climate change; the potential for litigation; and risks due to conflicts of interest.

There may be other factors than those identified that could cause actual actions, events or results to differ materially from those described in forward-looking information, there may be other factors that cause actions, events or results not to be anticipated, estimated or intended. There can be no assurance that forward-looking information will prove to be accurate, as actual results and future events could differ materially from those anticipated in such information. Accordingly, readers are cautioned not to place undue reliance on forward-looking information. Unless required by Canadian or United States securities legislation, the authors and Snowden Optiro undertake no obligation to update the forward-looking information if circumstances or opinions should change.

Energy Fuels Inc. Technical Report – Donald Rare Earths and Mineral Sands Project, Victoria, Australia

1 Summary

This Technical Report was prepared for Energy Fuels Inc. (Energy Fuels) to support the disclosure of Exploration Results, Mineral Resources and Mineral Reserves for Phase 1 of the Donald Rare Earths and Mineral Sands Project (Donald or the Property), a mineral exploration and development property located in western Victoria, Australia.

This Technical Report satisfies the requirements of Canadian National Instrument 43-101 Standards of Disclosure for Mineral Projects (NI 43-101) and the United States Securities and Exchange Commission's (SEC's) Modernized Property Disclosure Requirements for Mining Registrants as described in Subpart 229.1300 of Regulation S-K Disclosure by Registrants Engaged in Mining Operations (S-K 1300), and Item 601(b)(96) Technical Report Summary.

This Technical Report was authored by the following Qualified Persons (as such term is defined under NI 43-101 and S-K 1300):

• Mr. Allan Earl and Mrs. Christine Standing of Snowden Optiro, a business unit of Datamine Australia Pty Ltd (Snowden Optiro) were responsible for the preparation of this Technical Report, including the review of the geology, Mineral Resource estimates, mine planning, mining capital and operating cost estimates, and the economic analysis.

• Mr. Peter Allen of GR Engineering Services (GRE) was responsible for the review of the metallurgy, processing, infrastructure and processing capital and operating cost estimates.

• Ms. Gené Main and Mr. Peter Theron of Prime Resources (Prime) were responsible for the review of the environmental studies, permitting and social.

• Mr. Pier Federici of AMC Consultants Pty Ltd (AMC) was responsible for the review of the Mineral Reserve estimates.

The effective date of this Technical Report is 31 December 2025.

Unless otherwise specified, all units of currency are in Australian dollars ($) and all measurements are metric.

1.1 Property description, ownership and background

Donald is a planned greenfields mineral sands mine development in the Wimmera region of western Victoria, Australia. The Property comprises two areas, Mining Licence 5532 (MIN5532) and Retention Licence 2002 (RL2002), covering a combined area of 27,155 ha (271.55 km²), held by Astron Limited (formerly Astron Corporation Limited) (collectively Astron) through its subsidiary Donald Project Pty Ltd (DPPL).

Astron completed a "definitive feasibility study" in April 2023 for the Phase 1 operation within MIN5532 and a "pre-feasibility study" in June 2023 for the Phase 2 operation within RL2002, based on a plan to open pit mine the Donald heavy mineral (HM) deposit with on-site processing to produce a HM (zircon and titanium minerals) concentrate (HMC) and rare earth element concentrate (REEC) (Phase 1), and proposed future processing of the HMC into zircon and titania (titanium feedstock) final products (Phase 2).

On 4 June 2024, Energy Fuels entered into a farm-in and joint venture agreement and related ancillary agreements (Agreement) with Astron for a joint venture (JV) to develop the Donald deposit within MIN5532 and RL2002. Energy Fuels will contribute the first $183 million of equity capital to earn a 49% interest in DPPL. As at the effective date of this Technical Report, Energy Fuels held a 9.48% equity interest in DPPL and its remaining earn-in obligation is forecast to be $132 million (after taking into account funds already advanced and approximately $16 million of debt finance provided by EFRD which upon a final investment decision (FID) will be recognised as earn-in funding). Energy Fuels also entered into an offtake agreement for 100% of the Phase 1 and Phase 2 REEC monazite and xenotime production at commercial prices. Under the Agreement, and subject to its terms, Astron and its affiliates have the right to enter into an offtake agreement for 100% of the zircon and titanium HMC for processing at Astron's mineral separation plant in China and at third-party facilities.

Energy Fuels Inc. Technical Report – Donald Rare Earths and Mineral Sands Project, Victoria, Australia

DPPL completed an "Updated Economics Study" on 14 July 2025 to support a FID for the Phase 1 operation following the receipt of major regulatory approvals including the Work Plan covering the initial 19 years of operation (Phase 1A area of 1,143 ha in MIN5532). This study was updated in the "Draft - Donald Project Revised Economics Study Q4 2025" report dated 19 December 2025.

Once commissioned, the combined Phase 1A and 1B operation will mine and process 7.5 million tonnes per annum (Mt/a) of ore and is expected to produce an average of about 192 thousand tonnes per annum (kt/a) of HMC and about 7,100 tonnes per annum (t/a) of REEC for the 40-year project life.

1.2 Geological setting and mineralization

The Murray Basin is a low-lying, saucer-shaped intracratonic depression in southeastern Australia hosting thin, flat-lying Cainozoic sediments overlying deformed early Palaeozoic turbidites, volcanic and volcaniclastic rocks of the Lachlan Fold Belt. A succession of Tertiary freshwater, marine, coastal and continental sediments containing HM were deposited into the basin.

The Late Miocene to Late Pliocene Loxton Sand is the host sequence to all the known HM sand deposits in the Murray Basin. These deposits are of two principal types: the coarser-grained strandline occurrences and the finer-grained "WIM-style" accumulations. The strandline-style deposits occur along the seaward face of ancient shorelines and are the result of concentration and winnowing in a littoral environment. The WIM-style deposits, named after the Wimmera area of the Murray Basin, consist of a solitary or composite broad, lobate sheet-like body of highly sorted HM associated with fine grained, micaceous sand with considerable areal extent. These deposits are thought to represent accumulations formed below the active wave base in a near-shore environment, possibly representing the submarine equivalent of the strandline-style deposits. The WIM-style deposits are typically considerably larger in tonnage and lower in grade than the strandline deposits. The HM sand deposits are typically buried beneath Quaternary and Tertiary aged fluvial sediments.

The WIM-style HM sands in the Property are concentrated mainly within the lower units (LP2 and LP3) of the Loxton Sand, which ranges in thickness from 10 m to 15 m. HM concentrations decrease in grade towards the top of the fine-grained LP2 unit. A medium to coarse-grained sand unit (Loxton Sand LP1) overlies the fine-grained LP2 unit.

North-south trending, discrete higher-grade zones have formed within the greater Donald deposit, presenting a focus for the initial stages of the mining operation. To the west, the mineralization deepens and overburden increases. On the southern margins, the fine-grained, silty HM sand disperses in an east-west direction following silty clay units, which are interpreted as washout zones that tend to contain no HM.

The Loxton Sand is overlain by clays of the Pliocene to Holocene Shepparton Formation, which range in thickness from 5 m to 20 m. "Stringer" sands of the overlying Quaternary Woorinen Formation develop as discontinuous or meandering channels of up to 10 m in thickness within the Shepparton Formation clays. The drillhole geology shows that the top of the Loxton Sand at the Donald deposit is reached at a depth of around 9 m, depending on local topography such as sand dunes.

The HM sand deposit typically comprises the following minerals of economic interest:

• Zircon

• Rutile (and anatase)

• Leucoxene

• Ilmenite

• Monazite

• Xenotime.

Energy Fuels Inc. Technical Report – Donald Rare Earths and Mineral Sands Project, Victoria, Australia

Zircon is rich in the element zirconium. Rutile (and anatase), leucoxene and ilmenite contain titanium. Monazite and xenotime contain rare earth elements (REEs).

1.3 Exploration and drilling

The Property has been subject to several major evaluation campaigns by three companies: Conzinc Rio Tinto Australia (CRA) from 1985 to 1991; Zirtanium Ltd from 2002 to 2004 and Astron from 2010 to 2022 and in 2025. There has been no previous mine production within the Property.

To date, 847 holes for a total of 20,622.9 m have been drilled within MIN5532 and a further 805 holes for a total of 20,944 m within the surrounding RL2002. This includes 133 pre-production, grade control (GC) and 10 sonic holes drilled within MIN5532 during 2025 for a total of 3,637.5 m. All holes were vertical and orientated perpendicular to the sub-horizontal mineralized horizon. Most holes were drilled using reverse circulation aircore (AC) with 82 holes in MIN5532 drilled by the sonic method for verification of the AC drilling, geotechnical and metallurgical testwork. In addition, groundwater monitoring bores were drilled.

Some problems with recovery were experienced in the pre-2004 drilling with refinements subsequently made to the drilling equipment and technique to improve recovery. In the 2002 Zirtanium drilling program, poor HM% correlation with the CRA drilling was reported. These results were disregarded for the Mineral Resource estimate, with the holes used for geological interpretation only.

For the 2004, 2010 and 2015 drilling programs, the entire sample from each 1 m interval was collected at the drill rig with the samples dried before splitting to remove any uncertainty from the splitting of wet samples at the drill rig. In 2022, the 1 m samples were split at the drill rig using a rotary splitter with attention paid to cleaning and minimizing contamination under the supervision of an Astron representative. The sample splits were deemed acceptable and quality control data indicated that the data from the 2022 drill program is of a high quality and suitable for resource estimation. All GC samples collected in 2025 are from the entire 1 m interval and samples were split at the assay laboratory. Recovery factors were not applied to the data.

1.4 Sample preparation, analyses, and data verification

Definition of Area 1 and Area 2 was defined based on the various drilling programs for data analysis and grade estimation for the 2022 and subsequent 2025 Mineral Resource estimate. Only assay data from the 2022 drilling program were used for the MIN5532 (Phase 1) Mineral Resource estimate in Area 1. Assay data from the 2004, 2010 and 2015 drilling programs were used for resource estimation in Area 2. The area covered by the 2022 drilling is coded as Area 1 and encompasses 97% of the total area of MIN5532.

Data from the CRA drillholes were used for geological interpretation only, due to the historical nature of the data and inconsistencies in the size fractions and analytical methodologies with the data obtained in 2022. Analytical laboratories Western GeoLabs in Perth and Titanatek were used by Zirtanium; however, no details are available on their accreditation. The sample preparation and analytical process involved drying, crushing, screening, desliming and centrifugal heavy liquid separation of the +38 μm/-1 mm fraction for HM (%) calculation and mineralogy analysis (+38 µm/-90 µm fraction).

From 2010, Astron adopted similar sample preparation and analytical procedures to those used by Zirtanium at Western GeoLabs. Analytical work for the 2022 drill program was performed by independent laboratory Bureau Veritas at its Adelaide, South Australia facility (ISO/IEC 17025 (2017) accreditation). The sample preparation and analytical process was:

• Oven dry samples

• Break up clays in bag with mallet

• Rotary split 500 g for testwork with an additional 500 g split off every 28^th sample for the laboratory duplicate

• Soak overnight in a 1% tetrasodium pyrophosphate (TSPP) solution

Energy Fuels Inc. Technical Report – Donald Rare Earths and Mineral Sands Project, Victoria, Australia

• Wet screen at 20 µm, 250 µm and 1 mm to create an in-size sample of between 20 µm and 250 μm

• Dry and weigh in-size and oversize samples

• Rotary split off approximately 100 g for heavy liquid separation

• Centrifugal and heavy liquid separation using tetrabromoethane (TBE), >2.96 g/cm³ specific gravity (SG)

• Wash, dry and weigh sinks.

Total HM was measured in the +20 μm/-250 μm fraction and reported as a percentage of the whole sample. Mineralogy on 53 composite samples, generated from individual samples of >1% total HM from the 227 holes drilled in 2022 was also performed by Bureau Veritas using x-ray fluorescence spectrometry (XRF), laser ablation inductively coupled plasma mass spectrometry (ICP-MS) and QEMSCAN^® analysis.

Quality control data for the 2004 drill program included interlaboratory analysis of duplicate samples. The results were good, with correlation coefficients of over 0.94 for total HM, slimes and oversize. AMC reviewed the quality control data for laboratory repeat analysis of drill samples from the 2010 drilling and field duplicate and laboratory repeat analysis of drill samples from the 2015 drilling. AMC reported that the 2015 field duplicates showed a bias for total HM and oversize. For the MIN5532 (Phase 1) Mineral Resource, this data was used for estimation within Area 2 only.

Quality assurance and quality control (QAQC) procedures for the 2022 drilling program included insertion of standards and field duplicates at the drill site (rate of 1 in 40). Blank samples were not inserted and are generally not used in the mineral sands industry. In addition, duplicate and standard samples were inserted by the laboratory. Performance of the standard samples throughout the program was considered moderately acceptable and no bias was noted for HM, slimes or oversize contents over time. The field duplicate samples showed no overall bias for total HM and a small bias to lower slimes and oversize contents in the duplicate samples. The duplicate analysis indicates good overall precision. The results from internal laboratory standards and duplicates were reviewed by Snowden Optiro and the overall performance of the analyses was deemed acceptable. The Qualified Person concluded that the data from the 2022 drill program is of a high quality and suitable for resource estimation. There is less confidence in the assay data from the 2004, 2010 and 2015 drill programs, which was limited to resource estimation in Area 2.

No issues were noted by the Qualified Person from verification work completed on data management, surveying, sample security and sample analysis. There were no drilling or sampling programs in progress during the time of the Qualified Person's site visit in August 2024.

1.5 Mineral processing and metallurgical testwork

The finer grained (generally <90 μm) WIM-style HM historically were not suited to simple gravity separation and to achieve effective selectivity and recovery, attritioning has been required to remove the iron staining and clay cementation to present clean particle surfaces for a flotation stage.

The development of a new spiral in the late 1990s (the FM1 and subsequently the MG12) was effective in separating HM down to around the 20 μm particle size and pilot plant testwork on a bulk sample from a test pit at the Donald deposit in the mid-2000s was successful in producing a pre-concentrate. From 2010, ongoing testwork developed a gravity spiral and flotation flowsheet to produce an HMC.

To reduce flowsheet risk, a confirmatory testwork program using samples from a sonic drilling program commenced in 2015 for both the gravity circuit to produce HMC and downstream processing to final products. Continued re-assessment of the project from mining operations through the process flowsheet and product transport logistics, identified further opportunities for reducing capital, operating and product handling costs. The 2008 Environment Effects Statement (EES) conditions also prevented a mineral separation plant (MSP) on the project site, requiring the HMC to be transported elsewhere for sale or treatment.

A substantial body of testwork at pilot scale as well as additional testwork confirming the final flowsheet design and the response of ore samples extracted from the first few years of mining has been carried out since 2018. In 2019, a 1,000 metric tonne (t) bulk sample from the re-opened test pit was shipped to the Corridor Sands operation in Queensland where a pilot wet concentration plant (WCP) was constructed. After attritioning, the reprocessed HMC was floated in a continuous pilot flotation facility at an independent laboratory in Western Australia, which produced a quantity of on-spec rare earth concentrate grade at a high recovery. Further bulk samples from a sonic drilling program specifically targeting the first few years of mining were recovered and processed in 2022.

Energy Fuels Inc. Technical Report – Donald Rare Earths and Mineral Sands Project, Victoria, Australia

Characteristics of the HMC produced (which were all in line with previous results) were:

• An average particle size of 50 μm

• 94.3% HM assaying 33.1% TiO₂ with 17.9% ZrO₂ and 0.87% CeO₂

• An estimated radioactivity of 12 becquerels per gram (Bq/g), as expected from the elevated HM grade

• The mass balance produced slightly lower overall recoveries than the continuous and integrated pilot plant as this testwork was carried out without the ability to recycle between stages

• Concentrate upgrade plant (CUP) processing of the raw HMC to separate minerals containing REEs returned results in accordance with previous testwork

• The opportunity to further streamline the CUP flowsheet by eliminating gravity (tabling) upgrading of the rare earth flotation concentrate was identified.

The final flowsheet comprises:

• Scrubbing for disaggregation

• Ex-pit trommel at 10 mm prior to pumping to the WCP

• At the WCP, screen the slurry at 1 mm to reject residual oversize that has not been further deagglomerated in pumping and transport

• Single stage hydro-cyclone desliming to reject -20 μm slimes

• Retention in a Lyons Feed Control Unit (LCFU) surge bin

• Mass flow and density-controlled feed to a rougher, middlings scavenger and cleaner gravity spiral circuit using MG12 spirals

• Interstage fine screening at 250 μm ahead of the final recleaner stage using HG10i spirals to produce a raw HMC

• Selective flotation of the rare earth minerals in the raw HMC into a concentrate with filter cake drummed and containerized for transport

• Filtering, stockpiling and loading of the final high-grade, rare earth free HMC into half-height containers for transport.

Table 1.1 summarizes the metallurgical performance including stagewise recoveries of HM and the valuable components from the in-size HM fraction as well as target HM grades at each process stage and final product grades.

Table 1.1 Metallurgical performance summary

Energy Fuels Inc. Technical Report – Donald Rare Earths and Mineral Sands Project, Victoria, Australia

Source: Astron, 2023a
Notes: Assumes no oversize in raw HMC, HMC and REEC. Assumes no slimes in HMC and REEC.

1.6 Mineral Resource estimate

The Mineral Resource estimate, which is within MIN5532 and includes part of RL2002 (to the south of MIN5532), contains data from a total of 844 vertical AC drillholes (for a total of 20,648 m) and one Calweld drillhole (for a total of 19 m) drilled by CRA, Zirtanium and Astron. Assay data from 82 sonic drillholes (for a total of 1,948.9 m), drilled for bulk density, geotechnical and metallurgical testwork and verification of the AC drilling. Data from the pre-production AC GC holes drilled during 2025 were not used for the 2025 Mineral Resource estimate. Density data from sonic holes were used for tonnage estimation.

The Mineral Resource was estimated in 2022, and this model was updated in 2025 to include updated density data and estimates of rare earth oxides (REOs). In addition, the monazite was re-estimated using data from ICP-MS analysis rather than XRF data, as was used for the 2022 Mineral Resource estimate.

Geological information from all historical drilling campaigns was used to inform the geological interpretation for resource modelling. Sample assay data (including mineral assemblage data) derived from the 2022 drilling program were used for grade estimation within Area 1. Assay data from the 2004 drilling program (assayed using the +38 µm/-90 µm fraction) and data from the 2010 and 2015 drilling programs were also used for HM, slimes and oversize estimation in Area 2. Data from the 2004 drilling was also used for estimation of the mineral assemblage components within Area 2.

Four lithological surfaces were interpreted for the resource model (top of Loxton Sand, base of LP1, base of LP2, and base of LP3) using all available geological logging data. Surfaces were interpreted to define the top and base of the mineralization using a nominal 1% total HM cut-off grade from the total HM contained within the +20 μm/-250 μm fraction (following calibration of the data from the -38 μm and the +38 μm/-90 μm fractions to the -20 μm and +20 μm/-250 μm fractions within Area 2).

Data compositing was not required, as all the sample intervals with assay data were 1 m. A hard boundary was used for variography and grade estimation of total HM within the mineralized horizon in LP1, LP2 and LP3. Examination of the slimes and oversize data indicated a gradational boundary from the mineralized domains to the surrounding material within LP1 and LP2 and soft boundary conditions were used for variography and grade estimation of slimes and oversize within these domains.

The distributions of the total HM, slimes and oversize data within each geological unit and within the mineralized horizon are positively skewed; however, the total HM, slimes and oversize all have low coefficients of variation (less than 0.95). High-grade outliers are not present and so top cut grades (cap grades) were not applied.

Variogram analysis was undertaken using a normal scores transformation to determine the total HM, slimes and oversize continuity. Kriging neighbourhood analysis was carried out to optimize the block size, number of samples used for grade estimation, search ellipse dimensions and the block discretization.

Energy Fuels Inc. Technical Report – Donald Rare Earths and Mineral Sands Project, Victoria, Australia

The mineral assemblage data was attributed to each drillhole interval that was incorporated into the composite sample, and the data was coded using the wireframe surfaces for the mineralized horizon and each geological sequence. Along-strike and across-strike variograms were examined for rutile, leucoxene, ilmenite, zircon, monazite and xenotime.

Block grades for total HM were estimated using both ordinary kriging (OK) and inverse distance squared (ID²) techniques, and block grades for slimes and oversize were estimated using OK techniques. The OK HM estimate was used for Mineral Resource reporting and the ID² estimate was used for validation of the OK estimate. The block model has a parent block size of 100 mE by 200 mN by 1 mRL. The parent blocks were allowed to sub-cell down to 25 mE by 50 mN by 0.25 mRL to more accurately represent the geometry and volumes of the geological units and the mineralization horizon. Grade estimation was into the parent blocks and a three-pass search scheme was used.

Block grades for the mineral assemblage components (all titania mineral subdivisions, zircon, monazite, and xenotime) and TiO₂, ZrO₂+HfO₂ and REOs were estimated using inverse distance cubed (ID³) techniques and grade estimation was into the parent blocks.

The estimated grades in the resource model were validated by:

• Visual comparison of the drillholes and blocks

• Comparing the mean input grades with the estimated block grades

• Examining trend plots of the input data and estimated block grades by easting, northing and elevation slices.

Data from a total of 149 density samples, collected from test pits in 2005 (nuclear density measurements) and 2018 (sand replacement), from nuclear density measurements in 2024, and from sonic drill core samples in 2022, 2024 and 2025 were used to determine average density values for the Shepparton Formation and LP1, LP2 and LP3. These average density values were applied for tonnage estimation of the 2025 Mineral Resource.

The 2025 Mineral Resource estimate was classified into the Measured, Indicated and Inferred categories, taking into account data quality, data density, geological continuity, grade continuity and confidence in the estimation of HM content and mineral assemblage. Measured and Indicated Mineral Resources were defined Area 1 (covered by the 2022 drilling on a nominal spacing of 250 mE by 350 mN) and where the mineral assemblage was determined by QEMSCAN^®, XRF and ICP-MS analysis. The eastern area of LP1 (Domain 210 - outside of mineralized horizon) and the LP3 unit within the area of 2022 drilling were classified as Indicated. Domain 210 is thinner in the east and grade estimation was supported by sparser data compared to the western area. Inferred Resources were not defined in the area drilled in 2022. Mineral Resources were classified as Indicated and Inferred in Area 2 (outside of the 2022 drilling) as the historical nature of the data, and changes in the grain size and data calibration reduced confidence in the data used for estimation. The LP2 unit was classified as Indicated where mineral assemblage data was obtained from the 2004 drilling and was classified as Inferred where there was a lack of mineral assemblage data. Mineral Resources within LP1and LP3 were classified as Inferred in Area 2.

The 2025 Mineral Resource for the Donald deposit within MIN5532 exclusive of the Mineral Reserve reported above a cut-off grade of 1.0% total HM is summarized in Table 1.2.

The Donald 2025 Mineral Resources have been classified in accordance with the definitions for Mineral Resources in S-K 1300, which are consistent with the CIM Definition Standards for Mineral Resources & Mineral Reserves (the 2014 CIM Definition Standards) incorporated by reference in NI 43-101.

Energy Fuels Inc. Technical Report – Donald Rare Earths and Mineral Sands Project, Victoria, Australia

Table 1.2 Donald Mineral Resource exclusive of Mineral Reserves within MIN5532 as of 31 December 2025 (100% equity)

Notes:

• Mineral Resources are reported on a 100% basis. As at the effective date of this Technical Report, Energy Fuels held a 9.48% interest in the Property.
• Mineral Resources that are not Mineral Reserves do not have demonstrated economic viability.
• Measured and Indicated Mineral Resources that are within the Mineral Reserve outline have been excluded from the reported Mineral Resource. Inferred Mineral Resources within the Mineral Reserve outline are included in the reported remaining Mineral Resource
• The reference point for the Mineral Resources is in-situ without assumed recovery modifying factors.
• The MIN5532 Mineral Resource has been classified and reported in accordance with the 2014 CIM Definition Standards incorporated in NI 43-101 and S-K 1300 Definitions.
• Total HM is from within the +20 µm to -250 µm size fraction and is reported as a percentage of the total material. Slimes is the -20 µm fraction and oversize is the +1 mm fraction.
• Estimates of the mineral assemblage (zircon, ilmenite, rutile (including anatase), leucoxene, monazite and xenotime) are presented as percentages of the total HM component. Estimates of the oxide components (presented as percentages of the total HM component) are contained within the minerals and are not in addition to the minerals. The REOs (CeO₂, Y₂O₃, Pr₆O₁₁, Nd₂O₃, Dy₂O₃, Tb₄O₇) are a subset of the TREO.
• All tonnages and grades have been rounded to reflect the relative uncertainty of the estimate, thus the sum of columns may not equal.
• The Mineral Resource is reported within MIN5532 above a 1% HM cut-off grade within a RF100 pit shell identified using the geotechnical parameters, operating costs, metal prices and recoveries disclosed in Item 15.2.1.

1.7 Mining and Mineral Reserve estimates

MIN5532 will be mined using a conventional strip-mining method, designed as about 500 m wide strips separated by in-situ ore bunds between the strips. Each strip comprises a series of 500 m wide and 250 m long mining blocks separated by bunds constructed from overburden stripped from the active mining area. The mining blocks will be extracted in a progressive sequence within each strip, before shifting to a new strip. Wells will be used to dewater the active mining area and the active tailings cell.

Process tailings will be returned to tailings cells constructed in the void left behind the active mining block. A downstream embankment will be constructed between the active tailings block and active mining block. Waste overburden will be backfilled behind the active tailings cell and above consolidated tailings.

The mining contract will include topsoil and subsoil stripping, overburden stripping, ore mining to an in-pit mine upgrade plant (MUP) using bulldozer push, construction of the tailings cells, overburden backfilling and subsoil and topsoil replacement and contouring. The bulldozer push to a tracked MUP reduces reliance on truck haulage on potentially soft pit floors. Final rehabilitation will be carried out by other specialized contractors.

Energy Fuels Inc. Technical Report – Donald Rare Earths and Mineral Sands Project, Victoria, Australia

The 2025 Mineral Resource block model and 2025 GC model were used to generate the 2025 Mineral Reserve estimate. To accommodate the wall angles in the pit optimization process and to provide better resolution of the mining boundaries, the resource model was re-blocked to 25 mE by 25 mN by 1.0 mRL cell size. A slope angle of 20° and a 6% ore loss to account for in-situ bunds were applied. The mining costs used for the pit optimization were developed from contractor costs, and processing costs from the process design consultant. Processing recoveries were supplied by Metmac Services Pty Ltd. Product prices were supported by market analysis reports.

The pit optimization considered the Measured and Indicated Mineral Resource model blocks only within the MIN5532 boundary, with all Inferred and unclassified blocks treated as waste. The pit optimization generated a series of nested pit shells for a range of revenue factors (RFs) ranging from 10% (RF10) to 110% (RF110) of the base case prices. There was very little change in the optimization results beyond the RF60 pit shell because the economic pit was constrained by the MIN5532 boundary. The RF50 shell, which targets the higher-grade areas within the Work Plan area (Phase 1A), was selected for the initial mining area and the RF70 shell, which covers the entire MIN5532 area, was selected for the remaining MIN5532 mine life.

The cut-off for defining ore, has been increased to improve cash flow. The cut-off within the Work Plan Area was further increased to improve the initial cash flow and reduce payback. This has been achieved by raising the base of the mine and lowering the top of ore surface to exclude lower value material

The excavation was designed to exclude cultural and environmentally significant areas, the external tailings storage facility (TSF), the process plant footprint, roads and other support facilities. An offset of 100 m was used from the MIN5532 boundary to the crest of the closest pit excavation. The floor of the mine design was a surface created from the combined RF50 and RF70 shells.

The Measured Mineral Resource component was classified as a Proven Mineral Reserve and Indicated Mineral Resource was classified as a Probable Mineral Reserve. The Donald Mineral Reserve estimate within MIN5532 reported by AMC as of December 2025 is summarized in Table 1.3.

The information in this Technical Report that relates to the MIN5532 Mineral Reserve estimate is based on information compiled by Mr. Pier Federici and fairly represents this information. Mr. Federici is a Fellow of the Australasian Institute of Mining and Metallurgy and a full-time employee of AMC and is independent of DPPL, Astron and Energy Fuels. Mr. Federici has sufficient experience that is relevant to the style of mineralization and type of deposit under consideration and to the activity being undertaken to qualify as a Qualified Person as defined in NI 43-101 and S-K 1300.

The Mineral Reserve is in a granted Mining Licence in good standing with state and federal environmental approvals and an approved Cultural Heritage Management Plan (CHMP) in place over the Phase 1A Work Plan area within MIN5532. The Work Plan area covers 1,143.4 ha, of which Astron owns freehold titles over an area of 705 ha. The remaining freehold titles are contracted to DPPL with settlement scheduled at FID.

The Phase 1A Work Plan area will support mining for about 19 years. Mining over areas within MIN5532 outside the Work Plan area will require a CHMP and approval of a Work Plan amendment. There is an additional 1,646.6 ha of land within MIN5532 outside of the Work Plan area. DPPL will need to engage with those landowners to ensure appropriate access under MIN5532 is secured.

Energy Fuels Inc. Technical Report – Donald Rare Earths and Mineral Sands Project, Victoria, Australia

Table 1.3 Donald Mineral Reserve within MIN5532 as of 31 December 2025 (100% equity)

Source: AMC, 2025a
Notes:

• Mineral Reserves are reported on a 100% basis. As at the effective date of this Technical Report, Energy Fuels held a 9.48% interest in the Property.
• The Mineral Reserve is based on Measured and Indicated Mineral Resources contained within a practical mine design.
• Estimates of the mineral assemblage (zircon, monazite and xenotime) are presented as percentages of the total HM component. Estimates of the oxide components (presented as percentages of the total HM component) are contained within the minerals and are not in addition to the minerals. The REOs (CeO₂, Y₂O₃, Pr₆O₁₁, Nd₂O₃, Dy₂O₃, Tb₄O₇) are a subset of the TREO.
• The Mineral Reserve is reported by individual heavy mineral components for transparency of mineralogical composition and processing considerations.
• The reference point for the Mineral Reserve is in-situ with allowance for mining recovery.
• All tonnages and grades have been rounded to reflect the relative uncertainty of the estimate, thus the sum of columns may not equal.
• The nominal cut-off grade is 1.0% HM using the metal price, cost and recovery assumptions as disclosed in Item 15.2.1.
• The MIN5532 Mineral Reserve has been classified and reported in accordance with the 2014 CIM Definition Standards incorporated in NI 43-101 and S-K 1300 Definitions.

Energy Fuels Inc. Technical Report – Donald Rare Earths and Mineral Sands Project, Victoria, Australia

1.8 Processing methods and infrastructure

The Phase 1 process flowsheet includes the following units:

• A tracked MUP located in the pit

• Screens, deslime hydro-cyclones, thickening plant

• A wet concentration plant (WCP)

• A concentrate upgrade plant (CUP), including a REEC packing plant

• A HMC storage and loading plant.

The process plant was designed at a nominal annual plant throughput of 7.5 Mt.

Run-of-mine (ROM) ore extracted from each mining block will be bulldozed and fed to the in-pit MUP situated close to the crest of the active mining face. The MUP has been designed to scrub and screen the ROM ore before pumping to the WCP, which can be relocated as it moves along the active mining strip. The main process plant and ancillary facilities will be situated in the northwestern corner of MIN5532.

ROM screens at the front end of the WCP have been designed to remove coarse (+1 mm) gangue particles from the scrubbed and screened (-10 mm) ROM material pumped from the MUP. Deslime hydro-cyclones will be positioned above the WCP surge bin to remove fine slimes from the ore slurry prior to entering the surge bin and subsequent spiral circuit. The WCP will include a series of spirals to separate the HM from the screened and deslimed ore.

The CUP will separate the minerals containing rare earth elements from the raw HMC produced in the WCP. This will be achieved by attritioning the HMC to ensure that the surfaces of all minerals are sufficiently exposed prior to the flotation circuit used to collect the rare earth minerals into the REEC. The flotation process uses various chemical reagents so the rare earth minerals float to the surface of the cell with the froth, while the remaining HM sink to the bottom of the cell. The CUP building will be sealed to prevent interference from the environment (rain and wind) and to limit personnel access and time spent in proximity of the product which contains radioactive mineral particles.

Dewatered REEC will be loaded directly from the product bin into 2-tonne bulka bags and loaded into half-height lined shipping containers that meet Class 7 radioactive material transport requirements. Containers will be sealed, weighed, labelled, and placarded in accordance with International Atomic Energy Agency (IAEA) regulations for Class 7 transport and all other applicable regulatory requirements. The filled containers will be stored on site until dispatched to Energy Fuels, who will be responsible for final disposal of the container liners.

The HMC storage facility will be in a separate structure where the product will be pumped from the CUP, dried and loaded into custom-built half height shipping containers with a front-end loader.

Ancillary process infrastructure includes a reagents storage and dosing facility and a flocculant storage and preparation plant for the safe delivery, storage and use of reagents.

Other site infrastructure will include an external TSF immediately south of the proposed process plant with 12 months capacity to store tailings until sufficient in-pit void space is available, a power substation and network, raw water supply, water treatment plant and access roads. Other ancillary facilities include administrative buildings, first aid building, ablution blocks, change rooms, crib rooms, laboratory, workshops, storage facilities and loading and delivery bays.

Energy Fuels Inc. Technical Report – Donald Rare Earths and Mineral Sands Project, Victoria, Australia

1.9 Permitting, environmental and social

Following the 2008 EES, approval has been received under the Federal Environment Protection and Biodiversity Conservation Act 1999 (EPBC Act) in 2009 and renewed in 2018. The CHMP was approved for the Work Plan area in 2014 and the radiation licence obtained in 2015 and renewed in 2024. The HMC export license issued in 2016 has expired and a new permit is being sought to export the REEC (a permit is no longer required for the HMC owing to the changed composition).

DPPL's amended Phase 1A Work Plan, incorporating and addressing formal feedback and comments received from Earth Resources Regulator (ERR) Victoria and its referral agencies, was approved in June 2025. Other approvals in progress include infrastructure outside of MIN5532 relating to road upgrades, road decommissioning, and other secondary licences and permits for groundwater extraction and surface water capture. A Rehabilitation Plan was prepared in 2023, and DPPL is in the process of determining the closure cost estimate and required bond. Once determined, this will be submitted for approval with ERR. A Construction Rehabilitation Bond of $27 million has already been approved by the regulator.

The Phase 1 operations within MIN5532 cover arable, mixed-use freehold farming land. The Phase 1A Work Plan area encompasses 1,143.4 ha of land. DPPL currently owns freehold titles for a total of 705 ha within the Work Plan area. The remaining freehold titles are contracted to DPPL with settlement scheduled at FID.

In June 2022, DPPL established the Community Reference Group. Membership comprises 25 representatives of local community, business, agency stakeholders and DPPL. The Community Reference Group aims to facilitate information exchange from DPPL to stakeholders and to provide an avenue for community members to raise project-related issues. The Community Reference Group operates in an advisory capacity and does not hold regulatory authority.

1.10 Costs and economic analysis

The life of mine (LOM) capital cost and operating cost estimates for the 40-year mine life were developed to an AACE Class 2 level of accuracy (typically -15% to +15%) which meets the requirements for a feasibility study. The effective date of capital cost estimate is Q4 2025. The components of the pre-production capital cost estimate are summarized in Table 1.4.

Table 1.4 Pre-production capital cost estimate

Source: DPPL

Sustaining capital costs (Table 1.5) were developed on a bottom-up basis consistent with the defined mining and processing strategy and the level of study supporting the FID. The estimate includes ongoing capital required to sustain operations over the LOM, such as mining equipment replacement and rebuilds, progressive in-pit tailings cell and embankment construction, mobile equipment additions, plant and infrastructure upgrades, and ongoing landform rehabilitation works. Cost allowances were derived from first principles estimates, vendor budget quotations, and benchmarking against comparable mineral sands operations, and were scheduled in the LOM model based on asset lives, production schedules, and maintenance strategies. Sustaining capital excludes initial development and pre-production capital and was prepared in real terms at the FID base date for inclusion in the cash flow model supporting the Mineral Reserve estimate.

Energy Fuels Inc. Technical Report – Donald Rare Earths and Mineral Sands Project, Victoria, Australia

Table 1.5 LOM sustaining capital costs

Source: DPPL

The LOM operating cost estimate summarized in Table 1.6 was developed from a range of quotes and benchmarking performed against other similar projects in Australia.

Table 1.6 LOM operating cost estimate

Source: DPPL

Mining operating costs were developed by DPPL based on the detailed strip mining method defined for MIN5532 using unit costs provided by potential mining contractors submitted via tender. Costs were derived from first principles using estimated mining activities, including topsoil and subsoil stripping, overburden removal, ore mining and haulage to the process plant, construction of in-pit tailings cells, progressive backfilling, and landform rehabilitation. Unit rates reflect contractor-supplied pricing, equipment productivity assumptions, haulage profiles, material rehandling requirements, and allowances for operational delays.

Processing operating costs were developed by Mineral Technologies, based on the selected flowsheet, design throughput, reagent and consumable requirements, power and water consumption, and staffing levels. Costs were estimated from similar mineral sands operations and vendor quotations and scaled to the project operating parameters.

Tailings and residue management costs are included within the mining and processing cost estimates and reflect the in-pit tailings deposition strategy, embankment construction, dewatering, and progressive backfilling.

General and administrative (G&A) costs were estimated by DPPL and include site administration, technical services, environmental management, safety, and corporate overheads attributable to the operation.

Logistics, marketing and royalty costs were estimated by DPPL and include concentrate handling, transport, port charges, marketing costs, and applicable royalties. These costs were included in the cut-off grade determination and the LOM operating cost model.

All operating costs were prepared in real terms, at the FID base date, and were incorporated into the LOM cash flow model supporting the Mineral Reserve estimate.

There is a 2.75% government royalty and $30 million closure cost for MIN5532.

Energy Fuels Inc. Technical Report – Donald Rare Earths and Mineral Sands Project, Victoria, Australia

The financial model developed for the Phase 1 project summarizes the annualized LOM plan with inputs derived from detailed mine planning, mining and processing schedules, and capital and operating costs. The model aligns to the key inputs as described in this Technical Report that underpin the overall project plan and is based on Proven and Probable Mineral Reserves only.

The LOM plan assumes ore mining at 7.5 Mt/a feeding the MUP. The resultant rougher head feed is processed in the WCP at a rate of about 1,060 t/h at 7,200 h/a, producing an average of approximately 192 kt/a of HMC and 7,100 t/a of REEC over the 40-year project life. HMC and REEC production is higher in the first six years because mining is initially focused within the RF50 shell and approved Work Plan area, which targets higher value mineral assemblages, resulting in a higher recoverable mineral output at a constant plant throughput.

Figure 1.1 presents a summary of annual (calendar year) post-tax cash flows to 2067. Initial capital expenditure commences in 2026. Following the initial investment period, which results in a maximum negative cash flow of about $473 million in mid-2027, payback is achieved in 2034. Over the LOM, the project generates a cumulative post-tax cash flow of about $3,000 million, a pre-tax and post-tax NPV of about $800 million and $496 million respectively, at an 8% discount rate applied to quarterly cash flows, with an IRR of 16%.

Figure 1.1 Donald Phase 1 LOM cash flow summary (100% equity)

Source: Snowden Optiro

A sensitivity analysis of the pre-tax financial model NPV considered a variety of value drivers to arrive at discrete upside and downside value impacts for:

• Pricing for REEC using the high/low prices reported in Item 22.1.2

• Pricing for HMC using the high/low prices reported in Item 22.1.1

• Operating costs (±10%)

• TiO₂ recoveries (74.1% / 84.1%)

• ZrO₂ recoveries (89.5% / 95.5%)

• US$ exchange rate (0.6/0.7)

• Discount rate (9%/7%)

• Capital costs (±10%).

Energy Fuels Inc. Technical Report – Donald Rare Earths and Mineral Sands Project, Victoria, Australia

The most significant and material driver of project value is concentrate pricing (Figure 1.2).

Figure 1.2 LOM pre-tax NPV sensitivity analysis

Source: DPPL

1.11 Other relevant data and information

The June 2023 "pre-feasibility study" for the proposed Phase 2 development comprised:

• Duplication of the Phase 1 throughput with 7.5 Mt/a ROM material mined and processed within RL2002 to produce HMC and REEC

• Construction of a mineral separation plant (MSP) on MIN5532, sized to process the HMC equivalent of 15 Mt/a mined from both MIN5532 (Phase 1) and RL2002 (Phase 2), to separate the HMC into premium (ceramic) and secondary (chemical) grade zircon and final titania products.

The Phase 2 historical resource estimate for the extensions to the Donald deposit, outside of MIN5532 and contained within RL2002, was reported by AMC in 2016 based on data from 794 AC holes drilled by CRA, Zirtanium and Astron. All holes are vertical and the spacing varies from 125 mE by 450 mN to 500 mE by 500 mN and were analyzed for HM, slimes and oversize contents. Mineral assemblage data were obtained from composite samples from 348 drillholes on a spacing of approximately 200 mE by 450 mN. Adjustments were applied to the mineral assemblage data obtained prior to 2015 for ilmenite and rutile.

The RL2002 historical resource was classified and reported in accordance with the guidelines of the JORC Code (2012). The Qualified Person has not done sufficient work to classify the historical estimate as a current Mineral Resource in accordance with CIM Definition Standards for Mineral Resources & Mineral Reserves or S-K 1300 Definitions.

AMC also prepared the reserve estimate for Phase 2 using the 2016 historical resource estimate and studies completed on RL2002 by Astron, which included cost and price inputs, a strategic mine schedule and recovery rates. AMC updated the inputs and assumptions where appropriate, using external sources such as contractor prices, its in-house proprietary tool to estimate mining and operating costs from first principles, and experience with similar mining projects. The results were also compared with benchmarks. The basis of the 2023 reserve estimate and related assumptions were established to a ±25% level of accuracy. The methodology in determining the reserve estimate was similar to that adopted for the Phase 1 Mineral Reserve estimate.

The RL2002 reserve was initially classified in accordance with the guidelines of the JORC Code (2012). The Qualified Person has not done sufficient work to classify the historical estimate as a current Mineral Reserve and the estimate does not meet CIM Definition Standards for Mineral Resources & Mineral Reserves and S-K 1300 Definitions.

Energy Fuels Inc. Technical Report – Donald Rare Earths and Mineral Sands Project, Victoria, Australia

The following material issues have been identified by the Qualified Person responsible for the review of the reserve estimate regarding modifying factors that may materially affect the progress of Phase 2 and the future conversion of Mineral Resources to Mineral Reserves:

• The reserve is in a Retention Licence without the necessary and state and federal approvals and permits in place for mining, environmental, cultural and social issues.

• DPPL has limited freehold ownership over the surface of RL2002 outside of MIN5532. There is no guarantee that land can be purchased or accessed in a timely manner to allow production to proceed.

• The reserve is based on a study at ±25% accuracy completed in June 2023. The HM and REE concentrate prices, and capex and opex assumptions used in the study are subject to review to reflect current market conditions.

• The financing and timing for the commencement of the Phase 2 operation has yet to be determined.

1.12 Conclusions and recommendations

Key conclusions of this Technical Report include:

• The Phase 1 project (MIN5532) comprises 293 Mt of Proven and Probable Mineral Reserves at 4.5% total HM.

• The 40-year Phase 1 LOM plan involves a 7.5 Mt/a open pit mining operation using conventional strip mining. Ore will be processed through a MUP, WCP and CUP to produce HMC and REEC for sale under offtake agreements with Astron and Energy Fuels respectively.

• The first capital expenditure is in 2026. Following the initial investment period, which results in a maximum negative cash flow of about $473 million in mid-2027, payback is achieved in 2034. Over the LOM, the project generates a cumulative post-tax cash flow of about $3,000 million, a post-tax NPV of about $496 million at an 8% discount rate applied to annual cash flows, with an IRR of 16%. The financial model is highly sensitive to HMC and REEC pricing, making commodity price fluctuations a critical factor. The Phase 1A operation covering the first 19 years of the LOM plan has key approvals in place for the Work Plan area, including federal environmental permits, a CHMP and a radiation licence. Key outstanding critical path items include final land acquisitions and approval of the REEC export licence.

The future development of Phase 2 will be subject to the receipt of additional approvals, securing the required surface rights and completion of a feasibility study.

Key recommendations include:

• Improving Mineral Resource confidence through additional drilling and data calibration

• Optimizing tailings handling and pit sequencing to reduce costs and enhance mine efficiency

• Strengthening financial and cost management with refined estimates, bulk purchasing strategies and lease vs purchase evaluations

• Securing HMC offtake agreements and financing options to mitigate revenue uncertainty and attract investment

• Enhancing stakeholder engagement and permitting efforts for long-term project viability.

Energy Fuels Inc. Technical Report – Donald Rare Earths and Mineral Sands Project, Victoria, Australia

2 Introduction

2.1 Terms of reference

This Technical Report was prepared for Energy Fuels to support the disclosure of Exploration Results, Mineral Resources and Mineral Reserves for Phase 1 of the Donald Rare Earths and Mineral Sands Project, a mineral exploration and development property located in western Victoria, Australia.

This Technical Report satisfies the requirements of Canadian NI 43-101 and the SEC's Modernized Property Disclosure Requirements for Mining Registrants as described in Subpart 229.1300 of Regulation S-K Disclosure by Registrants Engaged in Mining Operations (S-K 1300), and Item 601 (b)(96) Technical Report Summary.

The Technical Report was authored by the following Qualified Persons (as such term is defined under NI 43-101 and S-K 1300):

• Mr. Allan Earl and Mrs. Christine Standing of Snowden Optiro, a business unit of Datamine Australia Pty Ltd (Snowden Optiro) were responsible for the preparation of this Technical Report, including the review of the geology, Mineral Resource estimates, mine planning, mining capital and operating cost estimates and the economic analysis.

• Mr. Peter Allen of GRE and an Associate of Snowden Optiro was responsible for the review of the metallurgy, processing, infrastructure and processing capital and operating cost estimates.

• Ms. Gené Main and Mr. Peter Theron of Prime Resources and Associates of Snowden Optiro were responsible for the review of the tailings dam, environmental studies, permitting and social.

• Mr. Pier Federici of AMC was responsible for the review of the Mineral Reserve estimates.

Mr. Earl completed a two-day site visit to the Property in August 2024. The site visit included an inspection of the proposed mining area, an inspection of the core shed, the collection of samples for check analysis, confirmation of a previous drillhole collar location (DM170) within rehabilitated agricultural land, a visit to nearby towns and an inspection of the likely service areas.

Mr. Pier Federici conducted a site visit to the Property in July 2013. The purpose of the visit was to familiarize himself with the site conditions, including existing mining activities, proposed pit limits, waste dump locations, site drainage and geotechnical considerations, access to the deposit, general landforms, and areas of vegetation proposed to be preserved. During the visit, Mr. Federici also observed sample preparation activities. In Mr. Federici's opinion, no material changes have occurred at the site since the time of the visit that would materially affect the Mineral Reserve estimate.

Site visits were not carried out by the other Qualified Persons, as there was no additional work or development completed at the Property that would contribute materially to the technical information and data provided. Mrs. Standing has previously conducted site visits to the WIM 150 and Avonbank WIM-style deposits in the Murray Basin.

All the Qualified Persons are eligible members in good standing of a recognized professional organization (RPO) within the mining industry and have at least five years of relevant experience in the type of mineralization and type of deposit under consideration and in the specific type of activity that the Qualified Person is undertaking as disclosed in Table 2.1 at the time this Technical Report was prepared.

Energy Fuels Inc. Technical Report – Donald Rare Earths and Mineral Sands Project, Victoria, Australia

Table 2.1 Responsibilities of each Qualified Person

Unless otherwise stated, the information and data contained in this Technical Report or used in its preparation was provided by the Property owner, Astron Limited (Astron). The Qualified Persons of this Technical Report reviewed information and documents provided by Astron via a virtual data room. The primary information sources were the "Donald Rare Earth & Mineral Sands Project Phase 1 - Definitive Feasibility Study" dated 27 April 2023, the "Donald Rare Earth & Mineral Sands Project Phase 2 - RL2002 Pre-feasibility Study Report" dated 26 June 2023, the "Donald Rare Earths & Mineral Sands Project Updated Economics Study" for Phase 1 dated 14 July 2025 and the "Draft - Donald Project Revised Economics Study Q4 2025" report for Phase 1 dated 19 December 2025. The virtual data room also included internal company technical reports, diagrams and maps, spreadsheets and reports prepared by Astron's external consultants.

Further information was received from the Astron representatives listed in Table 2.2 in response to queries submitted by Snowden Optiro.

Table 2.2 Astron information sources

The Property comprises two areas:

• Mining Licence 5532 (MIN5532) area where the licence holder is entitled to mine the land covered by the licence; explore for minerals and construct mining facilities related to the mining operation

• Retention Licence 2002 (RL2002) where a resource has been identified but the resource is not yet commercially viable to mine or is required to support an existing mining operation in the future.

Energy Fuels Inc. Technical Report – Donald Rare Earths and Mineral Sands Project, Victoria, Australia

The Donald Mineral Resources and Mineral Reserves with MIN5532 were initially classified under the 2012 edition of the Australasian Joint Ore Reserves Committee Code (JORC Code, 2012). The confidence categories assigned under the JORC Code (2012) were reconciled to the confidence categories in the Canadian Institute of Mining, Metallurgy and Petroleum (CIM) Definition Standards for Mineral Resources and Mineral Reserves (2014 CIM Definition Standards) and reported in compliance with NI 43-101 and
S-K 1300. For MIN5532, the confidence category definitions are the same and no modifications to the confidence categories were required. The RL2002 historical resource and reserve estimates disclosed in Item 24 were classified under the JORC Code (2012) and have not been adjusted to conform with NI 43-101 or S-K 1300. Energy Fuels is not treating the historical estimates as a current Mineral Resources or Mineral Reserves and are disclosed for background purposes only and should not be relied upon.

The Qualified Persons listed in Table 2.1 were responsible for this Technical Report and declare that they have taken all reasonable care to ensure that the information contained in this report is, to the best of their knowledge, in accordance with the facts and contains no material omissions.

In preparing this report, the Qualified Persons have extensively utilized information collated by other parties. The Qualified Persons have critically examined this information, made their own enquiries, and applied their general mineral industry competence.

The Qualified Persons believe that their opinions must be considered as a whole, and that selection of portions of the analysis or factors considered by them, without considering all factors and analyses together, could create a misleading view of the process underlying the opinions presented in this Technical Report. The preparation of a Technical Report is a complex process and does not lend itself to partial analysis or summary.

A draft copy of this Technical Report was provided to Astron and Energy Fuels for review on omission and factual accuracy. The Qualified Persons who have authored this Technical Report do not disclaim responsibility for the contents of this report.

The effective date of this Technical Report is 31 December 2025. As at the effective date of this Technical Report, none of the Qualified Persons had an association with Astron or Energy Fuels or their respective employees, or any interest in the securities of Astron or Energy Fuels or any other interests that could reasonably be regarded as capable of affecting their ability to give an independent unbiased opinion in relation to the Property.

This Technical Report constitutes a Feasibility Study for purposes of both NI 43-101 and S-K 1300 with respect to Donald Phase 1 and contains the initial NI 43-101 and S-K 1300 estimates of Mineral Resources and Mineral Reserves. This report does not include any NI 43-101 and S-K 1300 estimates of Mineral Resources or Mineral Reserves relating to Donald Phase 2.

Snowden Optiro, Prime, GRE and AMC will be paid a fee for the preparation by its Qualified Persons of this Technical Report based on a standard schedule of rates for professional services, plus any expenses incurred. This fee is not contingent on the outcome of the Technical Report, and neither Snowden Optiro nor the Qualified Persons will receive any other benefit for the preparation of this report.

2.2 Abbreviations and units

Unless otherwise specified, all units of currency are in Australian dollars ($) and all measurements are metric.

Energy Fuels Inc. Technical Report – Donald Rare Earths and Mineral Sands Project, Victoria, Australia

Table 2.3 Abbreviations and units of measurement

Energy Fuels Inc. Technical Report – Donald Rare Earths and Mineral Sands Project, Victoria, Australia

Energy Fuels Inc. Technical Report – Donald Rare Earths and Mineral Sands Project, Victoria, Australia

Energy Fuels Inc. Technical Report – Donald Rare Earths and Mineral Sands Project, Victoria, Australia

Energy Fuels Inc. Technical Report – Donald Rare Earths and Mineral Sands Project, Victoria, Australia

Energy Fuels Inc. Technical Report – Donald Rare Earths and Mineral Sands Project, Victoria, Australia

3 Reliance on information provided by the registrant

The Qualified Persons have relied on Energy Fuels for the legal aspects of land title and mineral tenure information, as summarized in Item 4 of this Technical Report, and the legality of any underlying agreement(s) that may exist concerning the permits or other agreement(s) between third parties, as summarized in Item 4 of this Technical Report. In particular, the Qualified Persons have relied on Energy Fuels' acceptance of information provided by representatives of Astron. The mineral tenure information was also confirmed on the GeoVic website of Resources Victoria¹.

The Qualified Persons have relied on Energy Fuels for guidance on applicable political and environmental matters outside the expertise of the Qualified Persons, as summarized in Item 20 of this Technical Report, and tax matters for the proposed Donald mining and processing operation, as summarized in Item 22 of this Technical Report.

Having made enquiries and taken appropriate steps to confirm this information in the public domain, the Qualified Persons consider it reasonable to rely on the information provided by Energy Fuels.

Energy Fuels Inc. Technical Report – Donald Rare Earths and Mineral Sands Project, Victoria, Australia

4 Property description and location

4.1 Location and area

Donald is in the Wimmera region of Victoria (latitude 36°29'25" S, longitude 142°46'21" E), approximately 300 km northwest of Victoria's capital city, Melbourne (Figure 4.1). The area of the Property is 271.55 km².

Figure 4.1 Location of Donald Property

Source: Astron

Energy Fuels Inc. Technical Report – Donald Rare Earths and Mineral Sands Project, Victoria, Australia

4.2 Type of mineral tenure

4.2.1 Legal framework

Resources Victoria sits within the Department of Energy, Environment and Climate Action (DEECA). It includes the Earth Resources Regulator (ERR) and Geological Survey of Victoria and plays a key role in:

• Regulating the resources industry to effectively manage risks to the environment and community

• Managing access to Victoria's resources for current and future use

• Policy development and regulatory reform

• Regulatory approval coordination

• Regional geoscientific investigations and data provision.

The Mineral Resources (Sustainable Development) Act 1990 (MRSD Act) is the legal framework for mining and extractive industries that is compatible with the economic, social and environmental objectives of the State. The Mineral Resources (Sustainable Development) (Mineral Industries) Regulations 2019 (MRSD Regulations) set clear work plan and rehabilitation plan requirements to manage risks associated with mining and mineral exploration.

An Exploration Licence (EL) gives the holder exclusive rights to explore for specific minerals within the specified licence area for five years. No mining activities can be undertaken on an EL. ELs may be renewed once, for up to five years. A second renewal, for up to five years, is only allowed in exceptional circumstances and where it can be demonstrated that there is a likelihood of the licensee identifying minerals during the period of the renewal. No further renewals are permitted.

Minimum expenditure conditions and progressive relinquishments of the licence area from year 2 (leaving 10% of the original licence area at the end of year 10) apply to an EL.

A Retention Licence (RL) is suitable where a resource has been identified but the resource is not yet commercially viable to mine or is required to support an existing mining operation in the future. The maximum term for a RL is 10 years and may be renewed for two additional periods of 10 years. The work required under a RL reflects the work program that was submitted with the licence application.

A Mining Licence (MIN) holder is entitled to mine the land covered by the licence, explore for minerals and construct mining facilities related to a mining operation. Before work can start, the holder needs to have an approved work plan for mining, provide a rehabilitation bond and obtain the necessary consents and permits.

A MIN can be granted for up to 20 years, or longer with the Minister's agreement, and there is no limit to the number of renewals, subject to the holder's record of compliance and various other matters, including whether mining will be feasible in the foreseeable future.

4.2.2 Property mineral titles

Astron's Donald Project comprises three granted mineral tenements and one application (Table 4.1 and Figure 4.1). MIN5532 and RL2002 form part of the Donald JV with Energy Fuels (the Property).

Table 4.1 Donald project mineral titles

Energy Fuels Inc. Technical Report – Donald Rare Earths and Mineral Sands Project, Victoria, Australia

*MIN5532 and RL2002 form part of the joint venture with Energy Fuels. Energy Fuels retains a Right of First Refusal on RL2003, as disclosed in Item 4.5.

Source: Astron

4.3 Issuer's interest

Astron holds its interest in the Donald JV mineral titles (MIN5532 and RL2002) through its subsidiary Donald Project Pty Ltd (DPPL, Figure 4.2).

Figure 4.2 Donald JV ownership structure

Source: Astron

On 4 June 2024, Energy Fuels entered into a farm-in and joint venture agreement and ancillary agreements (Agreement) with Astron for a JV to develop the Donald deposit within MIN5532 and RL2002. Energy Fuels will contribute the first $183 million of equity capital to earn a 49% interest in DPPL. As at the effective date of this Technical Report, Energy Fuels held a 9.48% equity interest in DPPL through its subsidiary company EFR Donald Ltd (EFRD). EFRD's remaining earn-in obligation is forecast to be $127.3 million (after taking into account funds already advanced and approximately $22.4 million of debt finance provided by EFRD, which upon FID will be recognised as earn-in funding).

Energy Fuels Inc. Technical Report – Donald Rare Earths and Mineral Sands Project, Victoria, Australia

Energy Fuels also entered into an offtake agreement for 100% of the Phase 1 and Phase 2 REEC monazite and xenotime production at commercial prices. Under the Agreement and subject to its terms, Astron and its affiliates retain a right to enter into an offtake agreement for 100% of the zircon and titanium HMC for processing at Astron's mineral separation plant in China and at third-party facilities.

Astron Mineral Sands Pty Ltd is the manager of the JV.

4.4 Surface rights

The Phase 1 operations within MIN5532 cover arable, mixed-use freehold farming land. The Phase 1A Work Plan area encompasses 1,143.4 ha of land (Figure 4.3).

DPPL currently owns freehold titles covering 705 ha within the Work Plan area, which are currently leased to local farmers for agricultural purposes. The remaining freehold titles within the Work Plan area are either held by Astron and leased to DPPL for the term of the Donald project, owned by DPPL, or are the subject of an option in favour of DPPL with settlement scheduled at FID. There is an additional 1,646.6 ha of land within MIN5532 outside of the Work Plan area (Phase 1B).

Figure 4.3 Plan of MIN5532

Source: Astron, 2025

Energy Fuels Inc. Technical Report – Donald Rare Earths and Mineral Sands Project, Victoria, Australia

Astron subsidiary, Jackson Mineral Sands Pty Ltd, owns an additional 1,138 ha of land outside of the Work Plan area, which is also leased out to local farmers providing alternative land for farming whilst the Work Plan area is mined, rehabilitated and returned to active farming.

Two parcels of Crown Land within the Work Plan area form part of the decommissioned open channel Wimmera Mallee Water Supply System and are no longer required for water supply purposes. Local roads within the Work Plan area are managed by Yarriambiack Shire Council and will revert to Crown Land if decommissioned.

A Public Conservation and Resource Zone runs along Dunmunkle Creek (to the west of the Work Plan area) and Richardson River (to the east of the Work Plan area) to protect and conserve the natural environment. Other areas of bushland to the northeast and southwest are conservation reserves designated Crown Land for reserve management.

Once the remaining freehold titles within the Work Plan area are purchased following exercise of the relevant purchase options by DPPL, the surface rights on MIN5532 will be sufficient for the proposed Phase 1A operation including all mining, processing and waste disposal. Overburden and process tailings will be backfilled into the void created by mining. A permanent ex-pit storage facility has been designed for tailings until sufficient mining void becomes available for backfilling.

4.5 Royalties, back-in rights, payments, agreements, encumbrances

The MRSD Act states that the royalty rate is 2.75% of the net market value (or mine gate value), which is the commercial value of the mineral at the time it is first sold, transferred or disposed of less any costs reasonably, necessarily, and directly incurred in connection with the sale, transfer or disposal, including insurance, freight and marketing.

The market value of the mineral means the value of the mineral if it were sold to an unrelated party in an arms-length commercial sale.

The Agreement with Astron provides for Energy Fuels to invest the first $183 million of capital required for the proposed Donald Project development and comprises:

• $1.5 million exclusivity fee, which has already been paid

• Funding for agreed project development activities until FID

• Secured interest free loans of up to $22.4 million for certain land and equipment acquisitions, whereupon following FID the loans will be recognised as earn-in funding and converted to equity in DPPL

• Sole funding of the balance of Donald project development costs up to a total of the earn-in amount of $183 million.

The conditions precedent under the Agreement included the transfer of assets, comprising the Donald deposit tenements (MIN5532 and RL2002) and water rights, to DPPL and Energy Fuels obtaining Foreign Investment Review Board (FIRB) approval from the Australian Government for its investment in Donald. FIRB approval was received on 19 September 2024, and completion of the JV agreement was achieved on 26 September 2024.

In addition, Energy Fuels agreed to issue common stock with a value of US$17.5 million to Astron in two tranches. The first tranche of US$3.5 million was issued upon the satisfaction (or waiver) of conditions precedent to the JV becoming effective. The second tranche of US$14.0 million will be issued upon approval of the FID for Phase 1 of the project.

Astron transferred MIN5532 and RL2002 and the water rights to DPPL during 2024. Land owned by Astron has been leased to the JV for the duration of operations with the JV responsible for all outgoings, rates and taxes. Astron will retain the right to develop the Jackson deposit on RL2003 independently. If the development of RL2003 is planned with a third party, Energy Fuels has a first right of refusal to participate.

Energy Fuels Inc. Technical Report – Donald Rare Earths and Mineral Sands Project, Victoria, Australia

Upon the expenditure of the full earn-in amount, Energy Fuels will have earned a 49% interest in the JV and Astron will retain a 51% interest. Any additional capital contributions will be funded by the parties pro-rata to their respective interests in the JV after taking into account the amount of certain pre-JV expenditure by Astron.

Energy Fuels' REEC offtake agreement will come into effect following the FID on the project. The price of REEC will be based on a formula derived from the market price of the constituent REOs (being neodymium and praseodymium combined, terbium and dysprosium), a payability factor and the actual assemblage of the REEC product (i.e. the percentage of the REOs). The JV will be responsible for organizing transport to Energy Fuels' White Mesa Mill in Utah, and the parties will work collectively in obtaining the necessary export permits.

The REEC offtake agreement will be subject to a floor price whereby, should the unit price of the REEC drop below the floor price, Energy Fuels may elect to suspend the REEC offtake until the realised prices of the downstream rare earth products recover. During this period, the JV may market the REEC product to third parties on the spot market.

Under the Agreement and subject to its terms, Astron and its affiliates have the right to enter into an offtake agreement for 100% of the project's HMC product on arms-length terms based on market pricing of the constituent products for processing at Astron's mineral separation plant (MSP) in Yingkou, China and at third-party facilities.

Astron Mineral Sands Pty Ltd as manager of the JV will charge a management fee of 5% of the allowable project costs pre-FID and 1.25% of allowable project costs post-FID plus reimbursement of all costs directly incurred by the manager in carrying out its duties as manager.

4.6 Environmental liabilities

Environmental bonds currently lodged against the Donald Project tenements are summarized in Table 4.1. A previously excavated test pit on land owned by Astron was fully rehabilitated in 2018. Prior to commencing Phase 1A site works, the bond requirements will be set by the responsible Victorian Government Minister.

A Rehabilitation Plan was prepared in accordance with the MRSD Act and associated MRSD Regulations in 2023 and approved as part of the Work Plan. A rehabilitation bond of $27 million to cover the liability up to process plant commissioning must be in place prior to commencing site works. Discussions with the ERR bonds team are ongoing regarding the bond calculation approach for the subsequent stages.

No other surface disturbances, land compensation, road maintenance or other liabilities or payments are required.

4.7 Permits

Following the 2008 Environment Effects Statement (EES), approval was received under the Federal Environment Protection and Biodiversity Conservation Act 1999 (EPBC Act) in 2009 and varied in 2018. The Cultural Heritage Management Plan (CHMP) was approved for the Work Plan area in 2014, and the radiation licence obtained in 2015 and varied and renewed in 2024. The HMC export licence issued in 2016 has expired and a new export licence is being sought relating to the REEC product.

DPPL's amended Work Plan, incorporating and addressing formal feedback and comments received from ERR and its referral agencies, was approved in June 2025. Other approvals in progress include infrastructure outside of MIN5532 relating to road upgrades, road decommissioning and water pipeline, and other secondary licences and permits for water supply connection, groundwater extraction and surface water capture.

Further details on the status of permitting are provided in Item 20.3.

Energy Fuels Inc. Technical Report – Donald Rare Earths and Mineral Sands Project, Victoria, Australia

4.8 Other significant factors and risks

The Work Plan area covers 1,143.4 ha, of which DPPL either owns or has obtained options to acquire the entirety of the Work Plan land area. There is an additional 1,646.6 ha of land within MIN5532 outside of the Work Plan area, of which 1,138 ha has already been purchased or contracted (Phase 1B).

DPPL will need to engage with the landowners within the Phase 1B area (to the extent they are different from the landowners within the Work Plan area and own land other than already held by DPPL) to ensure that appropriate access to the resource under MIN5532 is secured by the commencement of construction and preliminary mining activities for the Phase 1B operation. The options available to securing land access within the Phase 1B area include:

• Purchasing the freehold title or entering a long-term option to purchase agreement

• Leasing the land

• Entering into a compensation agreement under the MRSD Act

• Obtaining a compensation determination from the Victorian Civil and Administrative Tribunal (VCAT) or Supreme Court of Victoria (i.e. compulsory pathway).

There are Crown Land parcels and historical water channel reserves located both within the Work Plan area and the broader MIN5532. DPPL's understanding is that areas of Crown Land covered by MIN5532 are restricted to roads and de-commissioned water channels. DPPL will need to obtain the consent of the relevant Crown Land minister and other authorities, which cannot be unreasonably withheld, and can be granted subject to conditions including the payment of compensation.

There are nine sensitive receptors within 2 km of the MIN5532 boundary. Dust and noise modelling is currently being undertaken and will inform decision-making with respect to potential mitigation effects for these properties.

Energy Fuels Inc. Technical Report – Donald Rare Earths and Mineral Sands Project, Victoria, Australia

5 Accessibility, climate, local resources, infrastructure and physiography

5.1 Topography, elevation and vegetation

The topography of the Property is flat to gently undulating at an elevation of 126 m to 132 m above sea level. The land is extensively cleared of native vegetation (Eucalyptus and Buloke woodland and grassland) and used for cropping and livestock grazing.

5.2 Access

The Property can be accessed via sealed road from the capital city of Melbourne approximately 300 km to the southeast. Good access through the Property is provided by a network of unsealed roads and tracks servicing the freehold blocks.

5.3 Proximity to population centre and transport

The Work Plan area within the Property is 13.4 km east of the township of Minyip and approximately 65 km northeast of the regional centre of Horsham, in Victoria (Figure 4.1). The population of Minyip is approximately 390 and Horsham is approximately 15,600. There are several other townships in proximity to the Property, the largest of which is Donald with a population of approximately 1,400.

A rail line runs through Minyip. There are several aerodromes servicing nearby towns which are suitable for use by light aircraft.

5.4 Climate and length of operating season

The Property has a semi-arid climate with hot dry summers and cool wet winters, with most of the rain falling in winter and early spring. Mean diurnal temperature range from 4°C to 13°C during the winter months of June to August and from 13°C to 30°C during the summer months of December to February. The average annual rainfall is approximately 400 mm falling on an average of 98 days per annum.

The Phase 1 mine and process plant is scheduled to operate year-round on two 12-hour shifts per day, seven days per week with provision made for shutdowns and weather interruptions.

5.5 Infrastructure

The sufficiency of the surface rights to support the proposed Phase 1A project is discussed in Item 4.4.

Existing infrastructure includes telecommunications cables, low voltage transmission lines and water supply pipelines.

Power for the Phase 1 project will be supplied via an onsite hybrid microgrid using a mixture of solar, battery and diesel power generation.

The project will use a combination of groundwater, surface water and raw water supply for the mining and processing operations. The raw water supply will be drawn from Astron's Grampians Wimmera Mallee Water Headworks water allowance of 6.975 GL/a stored in Taylors Lake, outside of Horsham. An upgrade of the existing water reticulation systems to transmit the water from the Minyip Pumping Station to the mine site has been completed.

Access to the Property via the existing road network will be upgraded in part to meet local Shire and State government requirements.

Energy Fuels Inc. Technical Report – Donald Rare Earths and Mineral Sands Project, Victoria, Australia

5.6 Workforce

The project benefits from its location close to Horsham and Melbourne without the need for a fly-in-fly-out workforce. The preferred option for the Phase 1 operation is for a residential workforce to support the local communities. As such, the project is not planning to build permanent housing stock but rather work with local parties to jointly develop solutions, including utilizing existing housing stock in the area.

The construction workforce is estimated to peak at approximately 120 during an estimated 9-month construction period. During the operations phase, the 100-person residential workforce will work on two rosters:

• 2 weeks on/1 week off for shift roles

• 5 days on/2 days off for non-shift roles.

Energy Fuels Inc. Technical Report – Donald Rare Earths and Mineral Sands Project, Victoria, Australia

6 History

The Property has been subject to several major evaluation campaigns by three companies. The Donald deposit was discovered by CRA Exploration Ltd (CRA) in the early 1980s. Zirtanium Ltd (Zirtanium) acquired the tenements in 2000 and sold them to Astron in 2004. Since then, Astron has invested approximately $100 million in the project's development, including exploration, mining, metallurgical studies, obtaining necessary regulatory approvals and acquiring farmland and water rights. There has been no previous mine production within the Property.

6.1 CRA

During the 1980s, CRA actively explored the Murray Basin for multiple commodities, including brown coal, uranium, gold, diamonds and HM sands.

Anomalous accumulations of fine-grained HM were first identified through downhole geophysical logging (gamma ray spectroscopy). Following the discovery of the WIM 150 HM sands deposit near Horsham, CRA embarked on extensive regional AC drilling programs aimed at identifying similar mineral sands deposits. In 1990, CRA reported the partial delineation of the WIM 50, WIM 100, WIM 200 (current Jackson deposit) and WIM 250 (current Donald deposit within the Property).

From approximately 1982 to 1991, CRA undertook exploration and resource definition drilling over the Jackson and Donald deposits. CRA completed approximately 423 AC drillholes at Donald (within MIN5532 and RL2002) for an estimated total of 11,220 m (Table 6.1).

Table 6.1 Drilling by mineral title

Energy Fuels Inc. Technical Report – Donald Rare Earths and Mineral Sands Project, Victoria, Australia

Note: VHM - valuable heavy minerals.

6.2 Zirtanium

Between 2002 and 2004, Zirtanium Ltd completed 397 AC drillholes for a total of 6,097 m over the Donald deposit within MIN5532 and RL2002 (Table 6.1).

6.3 Astron

Astron completed four phases of AC drilling primarily focused on resource delineation between 2010 and 2022 and pre-production GC drilling in 2025 (Table 6.1):

• 2010 within MIN5532 and RL2002

• 2013 within RL2003 (Jackson deposit - not included in this report)

• 2015 within MIN5532, RL2002 and RL2003

• 2022 within MIN5532

• 2025 within MIN5532 (GC drilling for first two years of production).

Sonic drilling programs within MIN5532 were completed in 2015, 2022, 2024 and 2025 for geotechnical, bulk density and metallurgical testwork, to twin existing drillholes and add to the network of groundwater monitoring bores.

Further details of the drilling programs completed within MIN5532 and RL2002 are provided in Item 10.

A test pit was also excavated by Astron in 2005 about 1-2 km from the first mining blocks to a depth of 18 m. The test pit provided bulk samples for metallurgical testwork along with details on soil handling characteristics and observations of the rehabilitation requirements that will be encountered. A second bulk sample was taken, and the test pit was backfilled and rehabilitated during 2018.

6.4 Historical resource estimates

6.4.1 CRA

CRA completed resource modelling of the Donald deposit in 1990 (Smart and Allnut, 1990) and generated kriged block models for tonnage, grade and contained HM tonnes estimates. Separate tonnages for overlying clay and barren sand were also estimated.

The modelling was reportedly completed without a full assay database due to laboratory backlog and included assay results from two analytical methods. The older dataset used a minimum grain size of 38 μm and the newer dataset included grades for a concentrate nominally above 10 μm. Variogram analysis was used to determine grade estimation parameters, and the two datasets were subject to separate variogram analyses.

Energy Fuels Inc. Technical Report – Donald Rare Earths and Mineral Sands Project, Victoria, Australia

Drillholes with no mineralization were allocated mineralization thickness and grade values of zero and drillholes with suspect analytical data and abandoned holes were omitted from the analysis. The statistics indicated reasonable continuity structures for grade and thickness data in the north-south direction. A parent block size of 500 m by 500 m was used, selected from the drillhole spacing, total HM grades were estimated using OK, and a bulk density of 1.65 t/m³ was assigned for tonnage estimation.

The Donald deposit was reported as a north-northeast trending deposit of about 135 km² in area. The highest but thinnest HM grades were reported in the south. The bulk of the HM was found in the north-central zone. The reported tonnage-grade curve showed a slight inflection at the 3% HM cut-off and the "global resource" was reported as 1,270 Mt at 5.3% HM. This included the north-central zone of the Donald deposit which was reported as containing 520 Mt at 5.4% HM with a combined cut-off criteria of 4 m minimum thickness and 4% HM minimum grade.

CRA reported that the Donald resource estimate was classified as "Indicated" in accordance with the Australian Code for Reporting of Identified Mineral Resources and Ore Reserves (JORC Code,1989 edition). The Qualified Person has not done sufficient work to classify the historical estimate as a current Mineral Resource and the estimate does not meet CIM Definition Standards for Mineral Resources & Mineral Reserves and S-K 1300 Definitions. Energy Fuels is not treating the historical estimate as a current Mineral Resource and is disclosed for background purposes only and should not be relied upon.

CRA undertook further drilling after the resource study was completed in 1990.

6.4.2 Zirtanium

In February 2004, JB Mining remodelled the Donald deposit using all CRA's drilling results and results from Zirtanium's drilling carried out in August 2002 to verify the resource. The resource reported for Donald was 1,392 Mt at 6.4% HM (Thiess, 2004a).

A subset of the Donald deposit was remodelled by JB Mining in September 2004 and reported by Thiess, and mineral assemblage data was included in the resource estimate (Thiess, 2004b). The 2004 resource for the central area of Donald was reported to contain 355 Mt at 6.3% HM, and the HM fraction was reported to include 19% zircon, 20% leucoxene, 8% rutile and 30% ilmenite.

The resource for the Donald deposit was re-estimated in 2005. The mineralized horizon was interpreted using a 1.5% HM cut-off grade and a polygonal method was used for grade and tonnage estimation. Shepherd (2005) reported a resource of 187 Mt at 6.3% HM above a 1.5% HM cut-off and the HM fraction was reported to contain 19% zircon, 8% rutile, 20% leucoxene and 32% ilmenite.

The Qualified Person has not done sufficient work to classify the historical estimates as a current Mineral Resource and the estimates do not meet CIM Definition Standards for Mineral Resources & Mineral Reserves and S-K 1300 Definitions. Energy Fuels is not treating the historical estimates as a current Mineral Resource and are disclosed for background purposes only and should not be relied upon.

6.4.3 Astron

Historical resource estimates for the Donald deposit have previously been reported by AMC in 2006 (EL4433, including current MIN5532), 2010 (proposed MIN5532), 2011 (EL4433 and proposed MIN5532) and 2012 (EL4433 and MIN5532). The Qualified Person has not done sufficient work to classify the historical estimates as current Mineral Resources and the estimates do not meet CIM Definition Standards for Mineral Resources & Mineral Reserves and S-K 1300 Definitions. Energy Fuels is not treating the historical estimates as a current Mineral Resource and are disclosed for background purposes only and should not be relied upon. The most recent Mineral Resource estimate for MIN5532 (Phase 1) was prepared by Snowden Optiro and reported in 2022 and subsequently updated in 2025 to include updated density data and estimates of REOs, as disclosed in Item 14. The most recent Mineral Resource estimate for RL2002 (previously EL4433) was prepared by AMC and reported in 2016, as disclosed in Item 24.2 for Phase 2.

Energy Fuels Inc. Technical Report – Donald Rare Earths and Mineral Sands Project, Victoria, Australia

AMC prepared a resource estimate for EL4433 in 2006 which included the area that was subsequently covered by MIN5532 and extended into the area now covered by RL2002. The resource estimate was classified and reported in accordance with the guidelines of the 2004 edition of the JORC Code. AMC reported a total resource above a 1% HM cut-off grade of 693 Mt with an average grade of 5.1% HM, an average slimes content of 15.3% and an average oversize content of 7.3% (AMC, 2006). Mineral assemblage data was not available for the total resource and AMC reported a subset of the resource, with mineral assemblage data (Table 6.2). The total HM content of the subset resource was not reported, and the valuable heavy minerals (VHM) were reported as a percentage of the total material.

Table 6.2 Subset of the 2006 resource reported by AMC in EL4433 (which included MIN5532 and extended into RL2002)

Source: AMC, 2006

MIN5532 (Phase 1)

AMC prepared a resource estimate within the area of the proposed MIN5532 in 2010. This resource was classified and reported in accordance with the guidelines of the 2004 edition of the JORC Code. Mineral assemblage data was not available for the total resource and AMC reported a subset of the resource, with mineral assemblage data. The resource estimates for HM and the VHM assemblage, reported above a 1% total HM cut-off grade by AMC in 2010, are summarized in Table 6.3. Slimes and oversize contents were not reported for the subset resource. AMC reported an additional resource, without estimated VHM assemblage data, above a 1% HM cut-off grade of 235 Mt with an average grade of 2.8% HM, an average slimes content of 13.7% and an average oversize content of 14.3% (AMC, 2011).

Table 6.3 MIN5532 resource (subset with VHM) reported by AMC in 2010

Source: AMC, 2011

The resource estimate within the area of the proposed MIN5532 was updated by AMC in 2011 and was classified and reported in accordance with the guidelines of the 2004 edition of the JORC Code. Mineral assemblage data was not available for the total resource and AMC reported a subset of the resource, with mineral assemblage data. The resource estimates for HM and the VHM assemblage, reported above a 1% total HM cut-off grade by AMC in 2012, are summarized in Table 6.4. Slimes and oversize contents were not reported for the subset resource. AMC reported an additional resource, without estimated VHM assemblage data, above a 1% HM cut-off grade of 232 Mt with an average grade of 2.4% HM, an average slimes content of 14.0% and an average oversize content of 13.4% (AMC, 2012a).

Energy Fuels Inc. Technical Report – Donald Rare Earths and Mineral Sands Project, Victoria, Australia

Table 6.4 MIN5532 resource (subset with VHM) reported by AMC in June 2012

Source: AMC, 2012a

The resource estimate within MIN5532 was updated by AMC in September 2012. This resource was classified and reported in accordance with the guidelines of the current JORC Code (2012). AMC reported the resource above a 2.5% HM cut-off grade and between cut-off grades of 1% and 2.5% HM (AMC, 2012b). The resource estimates for HM and the VHM assemblage, reported above a 1% total HM cut-off grade are summarized in Table 6.5.

Table 6.5 MIN5532 resource (subset with VHM) reported by AMC in September 2012

Source: AMC, 2012b

AMC prepared a resource estimate within MIN5532 in April 2016. This was classified and reported in accordance with the guidelines of the JORC Code (2012). AMC reported a total resource above a 1% HM cut-off grade of 454 Mt with an average grade of 4.4% HM with an average slimes content of 14.2% and an average oversize content of 12.8% (AMC, 2016a). Mineral assemblage data was not available for the total resource and AMC reported a subset of the resource, with mineral assemblage data. The resource estimates for HM and the VHM assemblage, reported above a 1% total HM cut-off grade by AMC in 2016, are summarized in Table 6.6.

Table 6.6 MIN5532 resource (subset with VHM) reported by AMC in 2016

Source: AMC, 2016a

RL2002 (Phase 2)

AMC prepared a resource estimate within EL4433 (now RL2002) in 2011. This resource was classified and reported in accordance with the guidelines of the 2004 edition of the JORC Code. Mineral assemblage data was not available for the total resource and AMC reported a subset of the resource, with mineral assemblage data. The resource estimates for HM and the VHM assemblage, reported above a 1% total HM cut-off grade by AMC in 2012, are summarized in Table 6.7. Slimes and oversize contents were not reported for the subset resource. AMC reported an additional resource, without estimated VHM assemblage data, above a 1% HM cut-off grade of 1,521 Mt with an average grade of 3.8% HM, an average slimes content of 14.4% and an average oversize content of 8.2% (AMC, 2012a).

Energy Fuels Inc. Technical Report – Donald Rare Earths and Mineral Sands Project, Victoria, Australia

Table 6.7 RL4433 (now RL2002) resource (subset with VHM) reported by AMC in 2012

Source: AMC, 2012a

AMC prepared a resource estimate within EL4433 (now RL2002) which was classified and reported in 2012 in accordance with the guidelines of the JORC Code (2012). Mineral assemblage data was not available for the total resource and AMC reported a subset of the resource, with mineral assemblage data. The resource estimates for HM and the VHM assemblage, reported above a 1% total HM cut-off grade by AMC in 2012, are summarized in Table 6.8. Slimes and oversize contents were not reported for the subset resource. AMC reported an additional resource, without estimated VHM assemblage data, above a 1% HM cut-off grade of 1,203 Mt with an average grade of 4.0% HM, an average slimes content of 14.3% and an average oversize content of 7.0% (AMC, 2012c).

Table 6.8 RL4433 (now RL2002) resource (subset with VHM) reported by AMC in 2012

Source: AMC, 2012c

Energy Fuels Inc. Technical Report – Donald Rare Earths and Mineral Sands Project, Victoria, Australia

7 Geological setting and mineralization

7.1 Regional geology

The Murray Basin is a low-lying, saucer-shaped intracratonic depression in southeastern Australia hosting thin, flat-lying Cainozoic sediments. The basin overlies deformed early Palaeozoic turbidites, volcanic and volcaniclastic rocks of the Lachlan Fold Belt, is up to 600 m in thickness and extends approximately 850 km from east to west and 750 km from north to south, covering an area of approximately 320,000 km² over southwestern New South Wales, northwestern Victoria and southeastern South Australia (Figure 7.1).

Figure 7.1 Regional geological setting

Source: Astron

A succession of Tertiary freshwater, marine, coastal and continental sediments deposited HM into the basin. Much of the sedimentary sequence is the result of repeated marine incursions from the southwest, with the latest transgression-regression event resulting in deposition of the Late Miocene to Late Pliocene Loxton Sand (formerly called the Parilla Sand or Loxton-Parilla Sand). The Loxton Sand was deposited in shallow marine, littoral and fluvial conditions and comprises fine to coarse grained, commonly moderately well sorted sand with minor clay, silt, mica and gravel.

The Loxton Sand is the host sequence to all the known HM sand deposits in the Murray Basin. These deposits are of two principal types: the coarser-grained strandline occurrences and the finer-grained "WIM-style" accumulations.

The strandline-style deposits occur along the seaward face of ancient shorelines and are the result of concentration and winnowing in a littoral environment. These deposits are consistent with the present (and ancient) east and southwest Australian coastlines and are characterized by one or more relatively narrow composite lenses generally from 2 m to 12 m in thickness and are frequently persistent along any specific mineralized shoreline. These deposits are generally associated with relatively coarse, clean sand and gravel, consistent with any modern active beach environment.

Energy Fuels Inc. Technical Report – Donald Rare Earths and Mineral Sands Project, Victoria, Australia

The WIM-style deposits, named by CRA, consist of a solitary or composite broad, lobate sheet-like body of highly sorted HM associated with fine grained, micaceous sand with considerable aerial extent. These deposits are thought to represent accumulations formed below the active wave base in a near-shore environment, possibly representing the submarine equivalent of the strandline-style deposits. The WIM-style deposits are typically considerably larger in tonnage and lower in grade than the strandline deposits.

In the late Pliocene or early Pleistocene ages, the Murray Basin was closed by uplift in the southwest. Major lakes formed and deposited a thick sequence of sediments dominated by clay. The onset of arid climate conditions about 500,000 years ago added an extensive system of playa lakes and aeolian sands to the cover sequence of the central and northern Murray Basin. Quaternary to Recent river systems helped create the present-day surface geology and geomorphology. The HM sand deposits are typically buried beneath Quaternary and Tertiary aged fluvial and aeolian sediments.

7.2 Local geology and mineralization

The oldest rocks in the general area of the Property are a series of medium-grade (low to middle greenschist facies) metamorphic sediments of the Ordovician St Arnaud Group. These occur at depths ranging from 40 m to 75 m as determined by a few holes drilled into basement rocks (Figure 7.2).

Unconformably overlying the basement rocks are medium-grained sands (with occasional gravels) of the Eocene Renmark Group. These are essentially clean quartz sands with wood and seed fragments overlain by carbonaceous (occasionally lignitic) clays. The sands often contain crystalline masses or concretions of marcasite and are commonly water saturated. The Late Oligocene to Mid Miocene Geera Clay conformably overlies the Renmark Group as 10 m to 30 m thick sequence of dark green to black marly clay, with shell fragments and rare shark's teeth, and pyrite (which has the potential to form acid sulphate soil).

The HM sands in the Property are concentrated mainly within the lower units (LP2 and LP3) of the unconformably overlying Late Miocene to Late Pliocene Loxton Sand, which ranges in thickness from 10 m to 15 m. HM concentrations occur immediately above the Geera Clay and decrease in grade towards the top of the fine-grained LP2 unit. A medium to coarse-grained sand unit (Loxton Sand LP1) overlies the fine-grained LP2 unit: this unit also contains HM sands.

Minor amounts of iron oxides within the HM concentrations can form iron-cemented or indurated sandstone, developed as sub-horizontal layers within the deposit. The top of the HM deposit is often seen as a similarly developed cemented zone of less than 1 m in thickness. The HM occur as very fine laminae within clayey silt and have been shown to be gently dipping imbricated laminae within sub-horizontal bands of sediment, indicating deposition within an offshore deep-water, ripple bed environment.

North-south trending, discrete higher-grade zones have formed within the greater Donald deposit presenting a focus for the initial stages of the mining operation. To the west, the mineralization deepens and overburden increases. On the southern margins, the fine-grained, silty HM sand disperses in an east-west direction following silty clay units, which are interpreted as washout zones that tend to contain no HM.

The Loxton Sand is overlain by heavy (slaking) clays of the Pliocene to Holocene Shepparton Formation, which ranges in thickness from 5 m to 20 m. These widespread brown clays commonly show mottling due to hydrated iron oxides with local developments of haematitic pisolites or nodules. "Stringer" sands of the overlying Quaternary Woorinen Formation develop as discontinuous or meandering channels of up to 10 m in thickness within the Shepparton Formation clays.

Energy Fuels Inc. Technical Report – Donald Rare Earths and Mineral Sands Project, Victoria, Australia

Figure 7.2 Donald Project area generalized stratigraphic column

Source: Astron, 2023a

Note: Not to scale.

The drillhole geology shows that the top of the Loxton Sand at the Donald deposit is reached at a depth of around 9 m, depending on local topography such as sand dunes. The Loxton Sand is seen as a fine to medium grained sand at the base of the Shepparton Formation clays. A hard or silica cemented horizon (possibly the Karoonda Surface - a weathering profile created by a sea level hiatus) is seen a few metres into this sand, which can indicate the start of the very fine sand to silt zone of the Loxton Sand unit (LP2). The lower portion of the Loxton Sand is often below the water table.

Geological logging has differentiated the following interpreted depositional facies within the Loxton Sand (Figure 7.3). The depositional facies are used to geologically domain the deposit for resource modelling of the Donald deposit within MIN5532 (Phase 1). HM content is generally highest within the LP2 domain.

• LP1 - fine to very coarse friable quartz sands and minor silty, clay and gravel beds representing dunal, foreshore and surf zone sediments.

• LP2 - near-shore, very fine silty micaceous quartz sands, minor clays and gravels, representing sediments deposited below the wave base that show friable laminated and truncated HM mineralized beds. LP2 is the principal fine-grained HM target throughout the Murray Basin and contains most of the mineralization in the Donald deposit.

Energy Fuels Inc. Technical Report – Donald Rare Earths and Mineral Sands Project, Victoria, Australia

• LP3 - deep water sedimentation containing higher silt and clay material than LP2.

The Geera Clay is usually encountered at depths ranging from 18 m to 30 m and, due to its potential to form acid sulphate soil, mining is to be restricted to above this horizon.

Figure 7.3 Representative geological cross-section looking north along 5,959,750 mN with drillholes coloured by total HM%*

*Section location included in Figure 14.1

Note: As discussed in Item 14, for data and block model coding, the geological surfaces took precedence over the mineralization surfaces, with the mineralization constrained to within the Loxton Sand sequence (LP1, LP2 and LP3).

Source: Snowden Optiro

The mineralized horizon, within the Loxton Sand LP1, LP2 and LP3 layers, covers the entire area of MIN5532 (6 km east-west by 6 km north-south) and extends to the north, south, east and west of MIN5532 to within RL2002. Within RL2002 the mineralized horizon extends for 33 km north-south and from 3 km to 8 km east-west. The mineralized horizon ranges from 3 m to 20 m and has an average thickness of 9.8 m within MIN5532.

Energy Fuels Inc. Technical Report – Donald Rare Earths and Mineral Sands Project, Victoria, Australia

8 Deposit types

HM sand deposits contain concentrations of economically important minerals which are heavier than common sand minerals such as quartz. HM sand deposits typically comprise the following minerals of economic interest:

• Zircon

• Rutile (and anatase)

• Leucoxene

• Ilmenite

• Monazite

• Xenotime.

Zircon is rich in the element zirconium. Rutile (and anatase), leucoxene and ilmenite contain titanium. Monazite and xenotime contain REEs. Other minerals such as magnetite and garnet may also be present.

Victoria's HM sand deposits occur a long way from the modern coastline and reflect the presence of former inland seas and associated coastal processes. The two main types of HM sand deposits recognized in the Murray Basin are strandline deposits and Wimmera-style (WIM) deposits.

The strandline-style deposits occur along the seaward face of ancient shorelines and are the result of concentration and winnowing in a littoral environment. These deposits are consistent with the present (and ancient) east and southwest Australian coastlines and are characterized by one or more relatively narrow composite lenses, generally from 2 m to 12 m in thickness, and are frequently persistent along any specific mineralized shoreline for more than 2 km. These deposits typically have a 5-20% HM sand content with coarse-grained (>100 µm) HM assemblages associated with relatively coarse, clean sand and gravel, consistent with any modern active beach environment.

The WIM-style deposits, which are present and substantially pattern-drilled in the Property, are characterized by a sheet-like body of highly sorted HM associated with fine grained sand with considerable aerial extent. The deposits can be up to 25 m thick, 10 km long and 8 km wide and are relatively high tonnage and low grade (2-5% HM sand content) with a fine-grained (<250 µm) HM sand assemblage.

These deposits are thought to represent accumulations formed below the active wave base in a near-shore environment, possibly representing the submarine equivalent of the strandline-style deposits (Figure 8.1). The Donald deposit is geologically domained into LP1 beach facies, LP2 shallow sea and LP3 deeper water facies.

Energy Fuels Inc. Technical Report – Donald Rare Earths and Mineral Sands Project, Victoria, Australia

Figure 8.1 WIM-style HM sand deposit model

Source: Astron, 2023c

Energy Fuels Inc. Technical Report – Donald Rare Earths and Mineral Sands Project, Victoria, Australia

9 Exploration

As disclosed in Item 7, WIM-style HM sand deposits within the Property are typically buried beneath Quaternary and Tertiary aged fluvial and aeolian sediments, so surface exploration techniques such as mapping and geochemistry are not commonly performed or considered effective.

Excluding drilling, there has been no recent exploration conducted by DPPL relevant to the Property.

9.1 Geotechnical studies

ATC Williams Pty Ltd (ATC Williams) was engaged for the initial external tailings cells and subsequent in-pit tailings cell construction and deposition modelling. Numerous site geotechnical investigations have been undertaken since 2015 by Douglas Partners, GHD, and ATC Williams. ATC Williams provided geotechnical slopes of 1:2 (~27°) for in-situ slopes and 1:2.5 (~22°) for constructed slopes.

The most recent open pit geotechnical work completed by ATC Williams in 2022 was to obtain disturbed and undisturbed samples for laboratory testing to identify suitable material parameters for inclusion in the design of the co-disposal tailings facilities. The following testwork was completed by ATC Williams:

• Particle size distribution (PSD) of all material types

• Plasticity of fine-grained material encountered

• Emerson class testing to estimate dispersity of foundation material

• Particle density of the foundation materials

• Bulk density estimates from Lexan tube samples

• Compaction testing of remoulded samples for construction purposes

• Triaxial testing on selected undisturbed samples

• Remoulded permeabilities of foundation material for construction purposes.

Laboratory testing was undertaken to assess material strength of the Shepparton Clays, LP1, LP2 and Geera Clay materials. This work incorporated:

• Three consolidation tests on clay materials (Unit 2/Shepparton Formation, Unit 4/LP3 and Unit 5/Geera Clay) at Melbourne University

• Three triaxial tests.

In 2024 and 2025, ATC Williams completed an expanded program with:

• 17 geotechnical boreholes, 25 test pits, and 11 shallow boreholes in the pit, process plant, and external TSF areas

• 38 Lexan undisturbed samples and extensive in-situ testing (Pocket Penetrometer, Standard Penetration Test)

• Laboratory testing for PSD (sieve/hydrometer), Atterberg limits, SG, compaction, permeability, pinhole dispersion, Emerson, pH, California Bearing Ratio (standard and lime-stabilized), shrink-swell potential, sulphate content, and salinity/chemistry

• Integration of results into final pit slope design, TSF foundation design, and in-pit tailings consolidation modelling.

Energy Fuels Inc. Technical Report – Donald Rare Earths and Mineral Sands Project, Victoria, Australia

9.2 Hydrological studies

CDM Smith undertook a structured program of hydrogeological testwork to characterize surface and groundwater conditions at the Donald (MIN5532) Project. Data were acquired through a combination of drilling, bore construction, aquifer testing, and groundwater sampling using both slug and pump tests. Boreholes were completed to appropriate depths to intercept the key stratigraphic units, and monitoring wells were installed to enable long-term data collection. Groundwater levels were measured using dataloggers and manual dips to establish seasonal and short-term fluctuations. Sampling for water quality was conducted in accordance with industry-standard protocols to minimize contamination, and analyzed for major ions, trace elements, and nutrients at accredited laboratories, ensuring reliable baseline chemistry.

Hydraulic properties were characterized using constant rate aquifer pumping tests, step tests, and slug tests. These methods are widely accepted for determining transmissivity, storativity, and hydraulic conductivity. QAQC procedures included calibration of flow meters and water level loggers, appropriate test durations, and comparison of recovery data. Results demonstrated the presence of permeable zones within unconsolidated alluvial and sandy units, with hydraulic conductivities varying across the deposit. Pumping test responses confirmed aquifer yields sufficient for localized water supply, with storativity values indicative of confined to semi-confined behaviour in deeper units.

The conceptual model incorporates recharge from rainfall infiltration, limited lateral inflows, and discharge to surface features under natural conditions. A numerical groundwater model was developed and calibrated using observed heads, pumping test results, and long-term monitoring data. Material assumptions included homogeneity within mapped hydro-stratigraphic units, anisotropy aligned with depositional geometry, and steady-state recharge estimates derived from regional rainfall infiltration rates. Model outputs provided estimates of flow directions, aquifer connectivity, and sustainable abstraction rates, and were used to establish a water balance and predict potential drawdown impacts.

The Qualified Person concludes that the methods applied were appropriate and industry-standard, the data quality is sufficient for robust interpretation, and the models reliably characterize the aquifer systems underlying the project.

Energy Fuels Inc. Technical Report – Donald Rare Earths and Mineral Sands Project, Victoria, Australia

10 Drilling

10.1 Type and extent

Details of the drilling evaluating the Mineral Resource within MIN5532 are summarized in Table 10.1 and presented in Figure 10.1. Data from sonic holes, drilled for bulk density and geotechnical testwork and to expand network of groundwater monitoring bores, were not used for resource estimation. A representative cross-section is provided in Figure 7.3 with the location of the cross-section shown in Figure 14.1.

In addition, 133 AC holes and 10 sonic holes were drilled within MIN5532 during 2025 (Table 10.1 and Figure 10.2) that were not used for the current Mineral Resource estimate. A pre-production GC model was generated in 2025 using data from these drillholes.

Table 10.1 Drilling completed on MIN5532

*Not used for Mineral Resource estimation.

The drillholes completed over RL2002 (outside of MIN5532) are summarized in Table 10.2 with the drillhole collars shown in Figure 10.3.

All holes were vertical and orientated perpendicular to the sub-horizontal mineralized horizon.

Table 10.2 Drilling completed on RL2002 (outside of MIN5532)

Energy Fuels Inc. Technical Report – Donald Rare Earths and Mineral Sands Project, Victoria, Australia

Figure 10.1 Drilling in MIN5532 (black outline) coloured by year and used for Mineral Resource estimation

Source: Snowden Optiro

Figure 10.2 Drilling in MIN5532 (purple outline) during 2025 - not used for Mineral Resource estimation

Source: Astron

Energy Fuels Inc. Technical Report – Donald Rare Earths and Mineral Sands Project, Victoria, Australia

Figure 10.3 Extent of drilling in RL2002 (blue outline) coloured by program

Source: Astron

Energy Fuels Inc. Technical Report – Donald Rare Earths and Mineral Sands Project, Victoria, Australia

There are 19 holes drilled by CRA in the 1980s that intersected mineralization with >1% total HM within RL2002 to the north and outside of the extent of the MIN5532 Mineral Resource model (as discussed in item 14) and the RL2002 historical resource model defined by AMC (as discussed in Item 24.2). These drillhole intersections are listed in Table 10.3 and illustrated in Figure 10.4. All drillholes are vertical.

Table 10.3 Mineralised (>1% total HM) intersections within RL2002 not included in the extent of the resource models

Source: Snowden Optiro

Energy Fuels Inc. Technical Report – Donald Rare Earths and Mineral Sands Project, Victoria, Australia

Figure 10.4 Location of RL2002 and mineralized drillhole intersections outside of the MIN5532 and RL2002 resource models

Source: Snowden Optiro

Energy Fuels Inc. Technical Report – Donald Rare Earths and Mineral Sands Project, Victoria, Australia

10.1.1 Aircore drilling

The AC drilling completed by CRA over the Donald deposit was by a Mantis 200 drill rig. The drill spacing was generally at 500-800 m centres, with some closer spaced drilling (100-300 m) in selected areas. The average drillhole depth was 27 m below ground surface.

Zirtanium drilled the Donald deposit during 2002-2004 on a 250 m by 500 m grid pattern using a 10-tonne Mercedes truck-mounted (D28) Mantis 75 AC rig with a 300 cfm/175 psi compressor. NQ drill rods (70 mm diameter) with a three-blade drill bit were used for the 82 mm diameter holes. The holes were drilled to a depth of 24 m or until the Geera Clay was intersected.

The 2010 Astron program comprised infill AC drilling on MIN5532 and the area directly to the north. The purpose of this drilling was to:

• Provide sufficient coverage to enable the resource contained within MIN5532 to be upgraded to the Measured Resource category

• Obtain further information on the VHM composition within the HM resource and to enable the VHM resource to be increased

• Obtain further information on the resource to the immediate north where it appears that the high-grade zircon formation found within the resource area continues.

In 2015, a further AC drilling program was completed covering an area to the south of MIN5532 within RL2002. The program was primarily focused on delineating resources outside of MIN5532; however, a small number of holes provided information relevant to the southern part of MIN5532.

A Mantis 80 Toyota 6WD mounted drill rig with a NQ drill string and tungsten tip bladed NQ (82 mm) core bit was used for both programs.

Astron completed a further 245 vertical infill AC holes across MIN5532 in 2022 to:

• Provide new sample data for HM grade estimation covering the 20-250 μm size fraction to align with expected processing capabilities

• Address the difference in the domain sizes between the previously estimated and reported HM resource and VHM resource

• Sample more extensively above and below the deposit as previously interpreted

• Provide further logging information to enable a more detailed geological, domain-based resource estimation

• Determine rare earth mineral grades.

The drilling was completed on 250 mE by 250 mN and 250 mE by 500 mN patterns with holes spaced between existing drill traverses to improve the overall drillhole spacing. A Mantis AC75 drill rig with a 300 cfm/175 psi compressor and a NQ based drill string with three-blade AC drill bit was used for the program. All holes were drilled vertically, and the average hole depth was 26 m.

Astron undertook GC drilling in early 2025. For the GC program, AC drillholes were spaced on a 100 m by 100 m grid covering the first eight mining blocks, which represents approximately the first two years of mining production. Holes were drilled to intersect the entire Loxton Sand and to extent into the Geera Clay by 1 m. In addition to the regular grid-pattern, other drillholes included were:

• 14 drillholes infilling nearby areas that either were unavailable during the 2022 drilling or may become unavailable after the commencement of earthworks, due to the placement of long-term soil stockpiles or the need to establish drainage infrastructure.

Energy Fuels Inc. Technical Report – Donald Rare Earths and Mineral Sands Project, Victoria, Australia

• 10 drillholes drilled as twins of existing AC holes from the 2025 program with the only difference being that samples were split with a rotary splitter at the drill site. This exercise was undertaken to confirm if samples split at the rig can be considered reliably representative of the overall drill-sample.

• Four extra AC holes drilled north of the grid area following up on a potential high-grade zone seen in logging during the program.

10.1.2 Sonic drilling

In 2015, a 10-hole sonic drilling program was completed in the mining commencement zone of MIN5532. The core was sent to Mineral Technologies for metallurgical and engineering analysis. A selection of samples from were also analyzed by Ultra-Trace Laboratory for TiO₂, Fe₂O₃, Al₂O₃, ZrO₂, HfO₂ and loss-on-ignition by XRF analysis. No details are available on the accreditation or certification of Ultra-Trace Laboratory.

A second 25-hole program of sonic drilling was completed within MIN5532 during 2022 with the following objectives:

• To gather geotechnical information about the deposit including geology, material strength, compaction, moisture and bulk density

• To twin existing AC drillholes from the 2022 drilling program

• Collection of a bulk sample for further metallurgical testwork

• To add to the network of groundwater monitoring bores.

A 37-hole program of sonic drilling was completed during 2024, and a 10-hole program was completed in 2025 to collect additional geotechnical, bulk density and bulk samples for mineralogical analysis.

The 2015 and 2022 sonic drilling was carried out using a Boart Longyear LS600 drill rig. Sonic drilling after 2015 has been performed by Star Drilling Pty Ltd either using 6-inch core or 8-inch core. All geotechnical testwork was performed using a 6-inch core, whereas bulk samples were derived using the larger 8-inch core holes. Lexan Liner samples for bulk density analysis were taken from both hole sizes and from a range of locations across MIN5532. Assay data from the sonic holes were not used for Mineral Resource estimation.

10.1.3 Calweld drilling

In 2000, Zirtanium completed one large diameter (940 mm) Calweld drillhole to a total depth of 19 m to obtain a bulk sample.

10.2 Procedures

10.2.1 Surveying

The CRA reports do not include details of the survey methodology used for the holes drilled in the 1990s. As disclosed in Table 14.1, data from the CRA holes was used for geological interpretation only.

All Zirtanium's drillholes were surveyed using a global positioning system (GPS) unit with an accuracy of ±2 m. The drill collar elevation was estimated by GeoReality software using a high-resolution digital elevation model with an accuracy of ±1.2 m. AMC reported that later drillhole collars were surveyed using a differential GPS (AMC, 2016a).

The 2010 Astron holes were surveyed by GPS prior to drilling, and all holes were drilled within 1 m of the surveyed peg. At completion, the hole locations were surveyed by licensed surveyors.

All 2022 drillhole locations were set out using GPS survey equipment with the final hole locations also surveyed by licensed surveyors using a Leica Captivate GS18 unit and CS20 controller.

Energy Fuels Inc. Technical Report – Donald Rare Earths and Mineral Sands Project, Victoria, Australia

The coordinate systems used for all location data prior to 2025 was GDA94/MGA94 Zone 54 and the Australian Height Datum.

All 2025 drillhole locations were set out by licensed surveyors using a Leica Captivate GS18 unit and CS20 controller. Where holes were moved, the final hole locations also surveyed by licensed surveyors. The coordinate systems used for all 2025 location data was GDA2020/MGA2020 Zone 54 and the Australian Height Datum.

10.2.2 Sampling and logging

CRA

CRA's drill samples were collected at 1 m intervals in plastic lined garbage bins. Each drill sample was visually logged for lithology, colour, grainsize, sorting, grain shape, accessory minerals and bottom of the clay horizon. A handheld scintillometer recorded the gamma count of each sample to assess a relative HM grade.

Prior to 1988, drill samples were collected at 1 m intervals, with an equal dry split by weight composited for a 2 m interval and submitted for heavy liquid separation. A composite of the >2.96 g/cm³ SG concentrate for all drill samples from each hole was prepared for mineralogy studies.

The technique was modified in 1988 to reduce time spent on sample preparation. Previously, the work involved a conventional dry riffle splitting technique to generate a 1 kg dry sample representing the composite 1 m intervals in equal proportions by weight for a particular sample interval.

A concrete mixer was subsequently used to prepare a wet sample for a lithologically consistent bulked sample interval. This allowed ferruginous granules/pisolites in the drill samples to remain in suspension. CRA assumed that HM grains less than 100 μm would remain evenly disseminated in suspension. A wet sample was scooped from the mixer and dried to a 1 kg sample for analysis. A close correlation between the new technique and the conventional dry splitting technique was reported for recovered HM concentrates.

To allow for rapid analysis of samples collected during resource infill drilling, CRA used a hand coring technique to collect a sample from selected single metre intervals following compositing. Checks were made against the conventional dry splitting method and good correlations were reported.

On completion of the hole, a downhole natural gamma logging device was used to identify radiometric anomalies (thorium in monazite). Based on the gamma log values, individual samples were composited into larger intervals (usually two or three composites per hole) and sent for preparation and assay. The natural gamma logging results provided the depth to the top and base of mineralization and the interpreted thickness of the HM sand deposit in the drillhole. This information was later correlated with the assayed samples.

Zirtanium

For Zirtanium's drilling program, calico bags were used to collect the samples and water via a cyclone. At the completion of the drilling program, the geologist assessed the drill logs and selected the samples sent for HM analysis.

The drillholes were geologically logged at 1 m intervals recording the lithology, estimated HM, interpreted stratigraphic horizon and presence of indurated material. Samples for HM and size analysis were collected at 1 m intervals from identified mineralized horizon.

Prior to moving to the next location, each hole was plugged; backfilled and the site cleaned. The on-site geologist panned and extracted representative subsamples, and visually logged lithology, colour, estimated HM (%) and estimated HM grain size.

Energy Fuels Inc. Technical Report – Donald Rare Earths and Mineral Sands Project, Victoria, Australia

Astron

Astron's sampling method for the 2010 program replicated the process from Zirtanium's 2004 drilling campaign. Two types of samples were collected from each hole:

• Surface or topsoil sample using a pick and shovel

• HM samples collected at 1 m intervals from the identified mineralized horizon.

The HM samples were collected by cyclone at 1 m intervals. The recovered material was dropped into buckets from the cyclone and visually assessed for overall lithological description. Once potentially relevant HM-bearing sand and silt intervals were reached, a small sample of each interval was panned to determine the HM content, and the sample described in the logs. When the top of the HM-bearing interval was determined, calico sample bags were placed directly under the cyclone and the whole of the recovered material over 1 m was collected. The hole was terminated once the Geera Clay was intersected.

The drillholes were geologically logged at 1 m intervals recording the lithology, colour, mineralization, contamination and any other features.

For the 2022 program, representative samples were collected in calico bags at 1 m intervals using a rig mounted cyclone and rotary sample splitter (25% split). The holes were typically sampled from the top of the Loxton Sand until the intersection of the Geera Clay. The samples were sun dried before dispatch to the assay laboratory. Residue from the sample splitter was collected into larger calico bags for backup testing and recovery calculations. Field duplicate samples over a 1 m interval were also collected from the sample splitter every 40 samples. Standard samples sourced from Placer Consulting Pty Ltd were also inserted every 40 samples.

For the 2025 GC AC program, the entire sample for each 1 m interval was collected for 123 of the 133 holes, and samples were split at the assay laboratory. Ten twin AC holes were drilled, and a 25% split was taken at the drill rig with a rotary splitter. The 2025 sonic holes were sampled on 0.5 m intervals. All 2025 AC holes were sampled from the top of the Loxton Sand until the intersection of the Geera Clay. Geological logging recorded lithology, lithology proportion, grain size, colour, induration (presence and strength), hardness, geological stratigraphical unit and HM type and content estimation, and estimated clay content. The 2025 samples were not used for Mineral Resource estimation but were used to generate a separate GC model for the first two years of mining.

Sample representativity

While the AC drilling technique was extensively used for the evaluation of the Donald deposit, some problems with recovery were experienced in the pre-2004 drilling with refinements subsequently made to the drilling equipment and technique to improve recovery. In the 2002 Zirtanium drilling program, poor HM% correlation with the CRA drilling was reported. These results were not used for the current Mineral Resource estimate (as disclosed in Item 14), with the holes used for geological interpretation only.

For the 2004, 2010 and 2015 drilling programs, the entire sample from each 1 m interval was collected at the drill rig with the samples dried before splitting to remove any uncertainty from the splitting of wet samples at the drill rig. In 2022, the samples were split (25%) at the drill rig using a rotary splitter with attention paid to cleaning and minimizing contamination under the supervision of an Astron representative. The sample splits were deemed acceptable and quality control data (discussed in Item 11.3) indicated that the data from the 2022 drill program is of a high quality and suitable for resource estimation. Recovery factors were not applied to the data. For the 2025 GC AC program, the entire sample for each 1 m interval was collected for 123 of the 133 holes, and samples were split at the assay laboratory.

10.2.3 Data management and security

DPPL's samples are stored in a shed at Minyip.

Energy Fuels Inc. Technical Report – Donald Rare Earths and Mineral Sands Project, Victoria, Australia

Prior to February 2025, the DPPL master database was stored within a Microsoft Access format that contained all historical and recent drilling and assay information. The database was independently reviewed by AMC prior to its last Mineral Resource update in 2016 and by IHC Mining in 2021-2022.

Between 2022 and February 2025 all information added to the database and subsequent validation was the sole responsibility of Astron's senior geologist. Validation work included checking the imported data against the original assay reports. The database and backups were stored on Astron's Microsoft SharePoint document management platform, which has built-in access restrictions.

Since February 2025, DPPL's data has been managed in Quest by third party provider EarthSQL Pty Ltd in a fully relational SQL Server database hosted on Microsoft Azure. All data is gathered from either the field or the laboratory via pre-determined templates and vetted prior to being uploaded.

Energy Fuels Inc. Technical Report – Donald Rare Earths and Mineral Sands Project, Victoria, Australia

11 Sample preparation, analyses, and security

11.1 Sample preparation and analysis

11.1.1 CRA

CRA's samples were initially analyzed at independent laboratory AMDEL and at CRA's internal Belmont laboratory in Perth, Western Australia. No accreditation or certification details were disclosed for AMDEL or CRA's internal laboratory.

The analytical procedure used up to mid-1989 was heavy liquid and magnetic separation of a sieved fraction (>38 μm and <1 mm) followed by mineralogical detection by a combination of chemical and optical methods. Duplicate samples reportedly showed a good correlation between the two laboratories for total HM.

A new analytical technique was developed in 1989 (the 'WIM Method') to facilitate a more accurate assessment of the titanium mineral suite and to capture the fine material lost when sieving samples at 38 μm. The new procedure allowed better liberation of minerals cemented with clay or goethite, recovered the denser HM such as zircon and monazite in the <38 μm fraction and used chemical analysis rather than optical methods to determine the mineral assemblage of the HM fraction.

As disclosed in Table 14.1, data from the CRA holes were used for geological interpretation only. Assay data for HM, slimes, oversize and mineral assemblage data were not used for the MIN5532 Mineral Resource estimate due to the historical nature of the data and inconsistencies in the size fractions and analytical methodologies with the data obtained by Astron in 2022.

11.1.2 Zirtanium

Due to the large number of drill samples, multiple laboratories under the supervision of consultant's Titanatek were used by Zirtanium.

Titanatek gave each laboratory instructions for the sample preparation developed for WIM-style, fine-grained material. Independent commercial laboratory, Western GeoLabs laboratory in Perth, Western Australia performed the analytical work, using the following sample preparation procedure:

• Kiln dry samples (6-9 kg) at 110°C.

• Weigh samples for recovery estimate.

• Hammer each sample to break up clays.

• Screen sample using a 4 mm oversize screen, remove rocks and crush remainder of +4 mm fraction via a jaw crusher.

• Return crushed sample to 4 mm screen and hand crush +4 mm clays if required.

• Weigh and record all +4 mm product and discard.

• Riffle split all -4 mm material to obtain nominal 130 g subsample.

• Retain residue as reference/audit sample.

• Weigh subsample (25 g) and carry out soak/attrition desliming at 38 μm.

• Wet sieve to deslime and retain +38 μm product in aluminium tray and dry.

• Weigh dry sample and screen at +1 mm and record weight.

• Centrifugal heavy liquid separation of the +38 μm/-1 mm fraction.

• Screen sink material to +38 μm/-90 μm and weigh for in-size HM (%) calculation and mineralogy analysis.

Energy Fuels Inc. Technical Report – Donald Rare Earths and Mineral Sands Project, Victoria, Australia

• Composite mineral samples by weight for mineralogy analysis. Mineralogy testwork results were adjusted for potential use with +38 μm/-1 mm total HM assay data.

Comparative testwork using a centrifuge achieved an increase in the recovery of HM for the fine-grained fraction vs standard heavy liquid separation. Centrifuge was selected as the routine method for HM% determination using the following standardized procedures:

• Prepare a 25 g sample and place in a 110 ml centrifuge tube and add TBE (>2.96 g/cm³ SG) and mix.

• Centrifuge for 5 minutes at 1,500 rpm, re-stir and repeat spin.

• Scoop off floats and wash the sides of the tube with TBE and leave to settle for 5 minutes.

• Gently pour the remainder of the floats off, drain into filter paper and wash with TBE so that all HM are collected on the filter paper.

• Transfer to a wash basin and clean with acetone. Wash into a drying dish and oven dry.

• Weigh the HM content and save in labelled plastic bag.

Further testing by Titanatek confirmed that there was no statistical difference using a 25 g sample instead of the 70 g sample normally used by Titanatek. The smaller size was adopted and Western GeoLabs duplicated 1 in every 10 samples as part of its internal quality control procedure. To identify potential laboratory analytical errors, 1 in 20 samples was sent to external laboratories and results were statistically analyzed for bias, repeatability and reproducibility.

Analysis of the VHM component indicated a size range from 10 µm to 90 µm, the majority of which lies above 38 µm, with an average of approximately 50 µm to 55 µm. Western GeoLabs reported its HM% results as +38 μm/-1 mm, even though the majority of the VHM reported in the 38 μm to 75 μm range.

From 2004, Western GeoLabs reported the HM% results adjusted to a +38 μm/-90 μm HM size range definition for consistency with mineralogy assemblage testwork.

Geochempet Services performed mineralogical analysis on a sample set. To provide maximum control over the mineralogical data for resource estimation, every second hole on every second drill line was selected by Zirtanium for mineralogical analysis (1,000 m by 500 m coverage). The mineralogical analysis included composite samples prepared by Titanatek based on a 1.5% HM cut-off grade using a weighted average of the individual 1 m HM intervals.

No details are available on the accreditation or certification of Western GeoLabs, Geochempet Services or Titanatek.

11.1.3 Astron

From 2010, Astron adopted the same sample preparation and analytical techniques used by Zirtanium at Western GeoLabs. The only exception was the removal of jaw crushing +4 mm oversize material and the splitting of samples down to 250 g, which was outsourced to independent laboratory Gekko Pty Ltd in Victoria (ISO/IEC 17025 (2017) accreditation) to reduce the volume of material shipped to Western GeoLabs in Perth and improve the assay turnaround time.

As the plant process flowsheet did not include a crushing module, it was considered inappropriate to include this in the sample preparation procedure, resulting in a substantial increase in oversize percentage. Further analysis of the oversize by Western GeoLabs verified that most of this material is mineralized clay that has solidified during the sample drying process. Water in the trommel and subsequent processes will break this material down during processing, releasing any contained HM.

Western GeoLabs' laboratory in Perth carried out the heavy media separation by TBE and centrifuge.

Energy Fuels Inc. Technical Report – Donald Rare Earths and Mineral Sands Project, Victoria, Australia

Samples for mineralogy were either composited or selected on a per metre basis. Robbins Metallurgical (who purchased Titanatek and is now part of the Royal IHC group) prepared the samples and Geochempet Services performed the mineralogical studies.

Astron's analytical work during 2022 was performed by independent laboratory Bureau Veritas at its Adelaide, South Australia facility (ISO 9001:2015 accreditation). As disclosed in Item 14, Area 1 and Area 2 were defined, based on the drilling programs, for data analysis and grade estimation for the current Mineral Resource. The area covered by the 2022 drilling is coded as Area 1 and encompasses 97% of the total area of MIN5532.

The sample preparation and analytical process for the 2022 samples was:

• Oven dry samples

• Break up clays in bag with mallet

• Rotary split 500 g for testwork with an additional 500 g split off every 28^th sample for the laboratory duplicate

• Soak overnight in a 1% TSPP solution

• Wet screen at 20 µm, 250 µm and 1 mm to create an in-size sample of between 20 µm and 250 μm

• Dry and weigh in-size and oversize samples

• Rotary split off approximately 100 g for heavy liquid separation

• Centrifugal and heavy liquid separation using TBE (>2.96 g/cm³ SG)

• Centrifugal and heavy liquid separation of laboratory duplicate

• Centrifugal and heavy liquid separation of standard (supplied by Bureau Veritas) every 28^th primary sample

• Wash, dry and weigh sinks.

Total HM was measured in the +20 μm/-250 μm fraction and reported as a percentage of the whole sample.

The analytical testwork flowsheet is presented in Figure 11.1.

The need for additional breaking up of solidified clays and aggregate particles was monitored throughout the process with the possibility of adding a second pass of wet screening after further mechanical agitation in 1% TSPP solution, but this was deemed unnecessary at the time of the program.

Mineralogy on 53 composite samples, generated from individual samples of >1% total HM from 227 holes drilled in the 2022, was also performed by Bureau Veritas using XRF, laser ablation inductively coupled plasma mass spectrometry (ICP-MS) and QEMSCAN^® analysis. Testwork was performed on the in-size (+20 μm/-250 μm fraction) HM sinks composited by representative geological domains (LP1, LP2 and LP3) and geographic domains (from nearby or adjacent holes). This was done primarily to obtain enough sample mass for the QEMSCAN^® analysis.

The process for the mineralogy was:

• Compositing - homogenize and subsample as required

• QEMSCAN^® analysis

• Pulverise samples for assay

• XRF and ICP-MS

• Quantitative x-ray diffraction (XRD) analysis for validation and comparison with XRF and QEMSCAN^®.

Energy Fuels Inc. Technical Report – Donald Rare Earths and Mineral Sands Project, Victoria, Australia

Figure 11.1 2022 analytical testwork flowsheet

Source: Snowden Optiro, 2022

The "particle classification" data from the QEMSCAN^® analysis was used to estimate the titania minerals, XRF data was used to estimate the zircon, TiO₂% and ZrO₂+HfO₂%, and the ICP-MS data was used to estimate monazite, xenotime and REOs (including CeO₂% and Y₂O₃%) within the HM fraction.

The breakpoints used for the titania minerals were as follows:

• Rutile (including anatase) (≥95% TiO₂)

• Leucoxene (50-95% TiO₂)

• Ilmenite (30-50% TiO₂).

Zircon and monazite and xenotime were estimated using the follow conversions:

• Zircon = ZrO₂+HfO₂ / 0.667

• Monazite = CeO₂ / 0.28

• Xenotime = Y₂O₃ / 0.42.

The REEs (in parts per million (ppm)) were converted to REOs in percent using the following conversions, and TREO was calculated as the sum of these REOs:

• CeO₂ = Ce x 1.2284/10,000

• Dy2O₃ = Dy x 1.1477/10,000

• Er₂O₃= Er x 1.1435/10,000

• Eu₂O₃ = Eu x 1.1579/10,000

• Gd₂O₃ = Gd x 1.1526/10,000

Energy Fuels Inc. Technical Report – Donald Rare Earths and Mineral Sands Project, Victoria, Australia

• Ho₂O₃ = Ho x 1.1455/10,000

• La₂O₃ = La x 1.1728/10,000

• Lu₂O₃ = Lu x 1.1371/10,000

• Nd₂O₃ = Nd x 1.1664/10,000

• Pr₆O₁₁ = Pr x 1.2082/10,000

• Sm₂O₃ = Sm x 1.1596/10,000

• Tb₄O₇ = Tb x 1.1762/10,000

• Tm₂O₃ = Tm x 1.1421/10,000

• Y₂O₃ = Y x 1.2699/10,000

• Yb₂O₃ = Yb x 1.1387/10,000.

As discussed in Item 14, this data was used as input to an inverse distance estimate of the mineral assemblage components of the HM fraction.

The 2025 GC samples were analyzed by ALS Global Metallurgy in Perth, Western Australia (ISO 9001 accreditation). The following procedures were used for sample preparation and analysis:

• Clay dispersion in container of 1% TSPP solution - overnight soak time.

• Mild stirring for two minutes to disperse and homogenize.

• Dissociation by bottle-roll (steel).

• Wet screened to 1 mm, 250 µm and 20 µm using stacked screens.

• Sample fractions dried and weighted.

• Riffle split of +20 µm/-250 µm fraction for HM determination. Initially, 100 g was taken for the heavy liquid separation start weight, this was increased to 150 g mid-program to ensure enough sink material was produced for mineralogy work on low grade samples.

• Riffle spit fraction and processed via heavy liquid separation at 2.96 g/cm³ SG using TBE.

• The percentage of total HM calculated for the entire sample.

• Chemical analysis of heavy liquid separation sinks by XRF for mineral sands element suite and by D4ZM lithium metaborate fusion ICP-MS for REEs.

QEMSCAN analysis was not undertaken and so the individual titania minerals (rutile, leucoxene and ilmenite) were not determined. The above conversions for zircon, monazite, xenotime and REEs to REOs were applied.

11.2 Bulk density

11.2.1 CRA

CRA initially assigned bulk density values derived from the nearby WIM 150 deposit for the Donald (WIM 250) resource estimates due to their similarity. Initial determinations by CRA, derived from weighing a known volume of competent drill core, returned an average bulk density of 2.0 t/m³ within a range of 1.8 t/m³ to 2.2 t/m³. CRA used a more conservative bulk density of 1.75 t/m³ for the initial WIM 150 resource estimate.

Trenches excavated for bulk sampling allowed for further bulk density testwork, resulting in an average dry bulk density of 1.65 t/m³.

Energy Fuels Inc. Technical Report – Donald Rare Earths and Mineral Sands Project, Victoria, Australia

11.2.2 Zirtanium

Roadlab Pty Ltd, accredited by the National Association of Testing Authorities (NATA), conducted field density measurements from a test pit excavated during 2005 using a Troxler nuclear gauge. The results are summarized in Table 11.1 and confirmed the samples in Loxton Sand mineralization agreed with the bulk density value of 1.65 t/m³ used by CRA.

Table 11.1 Troxler nuclear gauge bulk density readings

Source: Astron, 2023a

11.2.3 Astron

As part of the 2022 sonic drilling program, 15 holes that were drilled primarily for geotechnical testwork were also used to collect samples for bulk density testwork specific to the geological domains (LP1, LP2 and LP3) used for the 2022 Mineral Resource estimate. Samples were analyzed by the ATC Williams Laboratory (ISO 9001 accreditation) using the Australian Standard test for Bulk Density (AS1289.6.4.1).

Further bulk density sampling was conducted as part of the 2024 and 2025 sonic drilling programs. These samples were analyzed using the same technique. Data for bulk density of soil types and the Shepparton Formation clays (0-10 m below surface) were also gathered from surface test pitting work as part of soil profile investigations in 2024.

Data from a total of 149 density samples, collected from test pits in 2005 (nuclear density measurements) and 2018 (sand replacement), from nuclear density measurements in 2024, and from sonic drill core samples in 2022, 2024 and 2025 were used to determine average density values for the Shepparton Formation and LP1, LP2 and LP3 (Table 11.2). These average density values were applied for tonnage estimation (Item 14).

Table 11.2 Bulk density testwork results

Source: Snowden Optiro, 2025b

11.3 Quality control quality assurance procedures

11.3.1 Pre-2022 analytical data

The CRA reports do not include details of QAQC data or procedures used for the samples collected in the 1990s. As disclosed in Table 14.1, data from the CRA holes was used for geological interpretation only in the MIN5532 Mineral Resource estimate.

A detailed review of the historical assay data by Astron (which included the collation of assay data from the original laboratory reports) revealed that different size fractions were used for the analysis of the total HM content of the whole sample post breakup and riffle splitting as follows:

• CRA: +38 μm/-1 mm fraction for HM%, mineralogy determined in +38 μm/-90 μm fraction (in-house laboratory) and +38 μm/-75 μm fraction (AMDEL) with all results adjusted to % of whole sample.

Energy Fuels Inc. Technical Report – Donald Rare Earths and Mineral Sands Project, Victoria, Australia

• Zirtanium, 2000 and 2002: +38 μm/-1 mm fraction for HM%, mineralogy determined in +38 μm/-90 μm fraction and adjusted to % of whole sample.

• Zirtanium, 2004: +38 μm/-1 mm fraction for HM%, mineralogy determined in +38 μm/-90 μm fraction and adjusted to % of whole sample.

• Astron, 2010 and 2015: +38 μm/-90 μm fraction.

Only assay data from the 2022 drilling program was used for the MIN5532 Mineral Resource estimate in Area 1. Assay data from the 2004, 2010 and 2015 drilling programs were used for resource estimation in Area 2 (Table 14.1).

Quality control data for the 2004 drilling program included interlaboratory analysis of duplicate samples. Western GeoLabs produced two 1 kg subsamples from 1 in 20 samples for analysis at Dune and Titanatek laboratories. The results were good with correlation coefficients of over 0.94 for total HM, slimes and oversize.

Six sets of twinned holes were drilled during 2004. Zirtanium (2005) reported that the mineralization intercept intervals were comparable, comparisons of the slimes and oversize data were acceptable, and the total HM content varied between the twin holes.

AMC reviewed the quality control data for laboratory repeat analysis of drill samples from the 2010 drilling and field duplicate and laboratory repeat analysis of drill samples from the 2015 drilling (AMC, 2016a). AMC reported that the 2015 field duplicates showed a bias for total HM and oversize. For the MIN5532 Mineral Resource estimate, this data was used for estimation within Area 2 only and was classified as Indicated at best (refer to Item 14.4).

11.3.2 2022 analytical data

QAQC procedures for the 2022 drilling program included insertion of standards and field duplicates at the drill site (rate of 1 in 40). Blank samples were not inserted and are generally not used in the mineral sands industry. In addition, duplicate and standard samples were inserted by the laboratory.

Standards

Prior to drilling, a batch of HM sand standard material was acquired from Placer Consulting Pty Ltd in Western Australia. The standard material was not officially certified. The standard samples used were primarily sourced for HM% content monitoring and the stated expected value for the supplied standard material was 3% HM.

Performance of the standard throughout the program was considered moderately acceptable averaging 2.87% HM over 121 standards assayed. The performance of the slimes content was more inconsistent with the content of the whole batch determined by large-scale mixing of slimes material into the batch. Similarly, the oversize (>1 mm) content was very low and consistent apart from two outlier results.

The standard could only be used to assess bias or drift in the batches of results over time. No bias was noted for HM, slimes or oversize contents over time.

For the 2022 drilling, the rates of insertion of standards are in line with industry standard rates. It is recommended that at least three different standards are included with the samples submitted for analysis.

Field duplicates

Duplicate samples were collected from the cyclone at the drill rig and inserted at a rate of 1 per 40 samples, with the results summarized in Table 11.3.

Assays for the original and repeat field duplicates show a high correlation (0.91 to 0.98). The 90^th percentile HARD (half absolute relative difference) values were good for total HM and slimes (17% and 10% respectively) and were acceptable for oversize (33%). The field duplicate samples showed no overall bias for total HM and a small bias (less than 4% relative difference) to lower slimes and oversize contents in the duplicate samples. The duplicate samples performed well and indicate good precision of the total HM, slimes and oversize analysis.

Energy Fuels Inc. Technical Report – Donald Rare Earths and Mineral Sands Project, Victoria, Australia

Table 11.3 QAQC results from field duplicate samples

Source: Snowden Optiro, 2022

Laboratory QAQC

Internal laboratory QAQC practices at Bureau Veritas for the 2022 drillhole samples consisted of standards and duplicates inserted at a rate of 1 in 28 throughout the program. The results of this work were reviewed by the Qualified Person for the Mineral Resource estimate, and the overall performance of the standard and duplicate analyses were deemed acceptable.

The average returned HM% value across 155 laboratory standards was 4.0% HM with an expected value of 4.0% HM. Scatter plots comparing laboratory duplicate results for HM %, slimes % and oversize % to the parent sample assays showed excellent correlation.

11.3.3 2025 analytical data

Field duplicates were not collected for the 2025 GC program, as the entire 1 m sample was submitted to the assay laboratory. Data from the twin-holes was used to assess the quality of the data. Only 17 field standards were submitted as part of the sample stream. The available data from the twin drillholes and standards have not indicated any issues. The 2025 samples were not used for Mineral Resource estimation but were used to generate a separate pre-production GC model for the first two years of mining.

11.4 Security

All samples from 2022 onwards were dispatched from site under the supervision of Astron's senior geologist. The samples were bagged into larger polyweave bags, numbered and zip tied prior to loading into numbered bulka bags on pallets for transport to Bureau Veritas using a commercial freight service.

Previously, assay data from the laboratory was emailed in spreadsheet files by batch number and imported into the Microsoft Access database by Astron's senior geologist. Since February 2025, data from the laboratory is emailed to EarthSQL Pty Ltd where it is vetted prior to being uploaded into the SQL Server database.

11.5 Qualified Person's opinion

Industry standard practice and a reputable laboratory have been used for sample preparation and sample analysis of the samples from the 2022 drilling program. Transfer of the 2022 assay data from the laboratory and inclusion in Astron's database was verified by the Qualified Person for the MIN5532 Mineral Resource estimate (refer to Item 12.4). No issues were noted by the Qualified Person on data management, sample security or sample analysis.

Energy Fuels Inc. Technical Report – Donald Rare Earths and Mineral Sands Project, Victoria, Australia

Following review of the data quality and assay techniques, the Qualified Person concluded that the dataset from the 2022 drill program is of a high quality and suitable for resource estimation and potential Measured classification. Within MIN5532 (Phase 1), only the 2022 assay data were used for estimation of 92% of the Mineral Resources: this includes all of the Measured Resources and 82% of the Indicated Resources in MIN5532 (refer to Item 14.4).

There is less confidence in the assay data from the 2004, 2010 and 2015 drilling programs, which were used for resource estimation in Area 2. Consequently, the Qualified Person assigned a lower classification, and Mineral Resources within Area 2 were classified as Indicated at best (refer to Item 14.4). Analysis of the CRA data and Zirtanium 2000 and 2002 samples used different size fractions, so assay data from these programs were not used for the current Mineral Resource estimate. Data from these drilling programs were only used to guide geological interpretation.

Energy Fuels Inc. Technical Report – Donald Rare Earths and Mineral Sands Project, Victoria, Australia

12 Data verification

12.1 Data management

The Qualified Person for the Mineral Resource estimate compared the assay data sheets for the 2022 drilling program from the laboratory against the assay data files provided by Astron. All the 2022 assay data extracted from the database matched the data in the assay data sheets supplied by Bureau Veritas. Only assay data from the 2022 drilling program were used for resource estimation in Area 1 (refer to Item 14). The area covered by the 2022 drilling is coded as Area 1 and encompasses 97% of the total area of MIN5532.

12.2 Survey

The Qualified Person checked the drillhole collar data for all 2000-2022 drillholes against the topographical surface provided by DPPL. A few minor discrepancies were noted between the topographical surface and the 2022 drillhole collars and, as the 2022 drillhole collars were surveyed by licensed surveyors (refer to Item 10.2.1), the topographical surface was adjusted to align with all 2022 drillhole collars.

During the site visit by the Qualified Person, confirmation of a drillhole collar location within rehabilitated agricultural land was completed.

12.3 Drilling and sampling

There were no drilling or sampling programs in progress during the time of the Qualified Person's site visit in August 2024.

12.4 Sample analysis

During the Qualified Person's site visit in August 2024, 21 samples from AC drillhole DMS170, completed during the 2022 drilling program, were selected and submitted to Bureau Veritas for analysis. The results are summarized as downhole plots of the total HM, slimes and oversize data in Figure 12.1. Correlation coefficients of the datasets are moderately high for total HM (r² = 0.84) and slimes (r² = 0.81) and high for oversize (r² = 0.94). The average total HM, slimes and oversize contents are slightly higher for the original assay data (Table 12.1). In the Qualified Person's opinion, the repeat analysis results indicate acceptable verification of the original 2022 assay data.

Table 12.1 Summary statistics from verification analysis of hole DMS170

Source: Snowden Optiro